WO2001014910A2 - Method for developing a system for identifying the presence and orientation of an object in a vehicle - Google Patents
Method for developing a system for identifying the presence and orientation of an object in a vehicle Download PDFInfo
- Publication number
- WO2001014910A2 WO2001014910A2 PCT/US2000/014903 US0014903W WO0114910A2 WO 2001014910 A2 WO2001014910 A2 WO 2001014910A2 US 0014903 W US0014903 W US 0014903W WO 0114910 A2 WO0114910 A2 WO 0114910A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- seat
- algorithm
- transducers
- database
- Prior art date
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/01516—Passenger detection systems using force or pressure sensing means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/01516—Passenger detection systems using force or pressure sensing means
- B60R21/0152—Passenger detection systems using force or pressure sensing means using strain gauges
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/0153—Passenger detection systems using field detection presence sensors
- B60R21/01532—Passenger detection systems using field detection presence sensors using electric or capacitive field sensors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/0153—Passenger detection systems using field detection presence sensors
- B60R21/01534—Passenger detection systems using field detection presence sensors using electromagneticwaves, e.g. infrared
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/0153—Passenger detection systems using field detection presence sensors
- B60R21/01536—Passenger detection systems using field detection presence sensors using ultrasonic waves
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/01544—Passenger detection systems detecting seat belt parameters, e.g. length, tension or height-adjustment
- B60R21/01546—Passenger detection systems detecting seat belt parameters, e.g. length, tension or height-adjustment using belt buckle sensors
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01554—Seat position sensors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
- G01S15/04—Systems determining presence of a target
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
- G01S15/06—Systems determining the position data of a target
- G01S15/42—Simultaneous measurement of distance and other co-ordinates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R21/00—Arrangements or fittings on vehicles for protecting or preventing injuries to occupants or pedestrians in case of accidents or other traffic risks
- B60R21/01—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents
- B60R21/015—Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents including means for detecting the presence or position of passengers, passenger seats or child seats, and the related safety parameters therefor, e.g. speed or timing of airbag inflation in relation to occupant position or seat belt use
- B60R21/01512—Passenger detection systems
- B60R21/01544—Passenger detection systems detecting seat belt parameters, e.g. length, tension or height-adjustment
- B60R21/01548—Passenger detection systems detecting seat belt parameters, e.g. length, tension or height-adjustment sensing the amount of belt winded on retractor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/88—Sonar systems specially adapted for specific applications
Definitions
- the present invention relates generally to the fields of sensing, detecting, monitoring and/or identifying various objects, and parts thereof, which are located within the passenger compartment of a motor vehicle.
- the present invention relates to an efficient and highh reliable method for developing a system for evaluating the occupancy of a vehicle by detecting the presence and optional! ⁇ orientation of objects in the seats of the passenger compartment, e.g.. a rear facing child seat (RFCS ) situated in the passenger compartment in a location where it interact with a deploying occupant protection apparatus, such as an airbag. and/or for detecting an out-of-position occupant.
- RFCS rear facing child seat
- the resulting system permits the control and selective suppression of deployment of the occupant protection apparatus when the deployment may result in greater injury to the occupant than the crash forces themselves. This is accomplished in part through a specific placement of transducers of the system, the use of a pattern recognition system, possibl a trained neural network, and/or a no ⁇ el analysis of the signals from the transducers.
- Applications engineering of occupant sensors comprises, inter alia, determining the location of the transducers, designing the transducer holders, determining the wiring layout, performing a tolerance study on the transducer locations and angular orientation, designing the circuits for the particular vehicle model, interfacing or integrating the circuits into the vehicle electronic system, and adapting the occupant sensor system to the particular vehicle model.
- Inflators now exist which will adjust the amount of gas flowing to or from the airbag to account for the size and position of the occupant and for the severity of the accident.
- the vehicle identification and monitoring system discussed in U.S. Pat. Nos. 5,829,782 and 5.943.295 among others, will control such inflators based on the presence and position of vehicle occupants or of a rear facing child seat.
- the instant invention is concerned with the process of adapting the vehicle interior monitoring systems to a particular vehicle model and achieving a high system accuracy and reliability as discussed in greater detail below.
- the return ultrasonic echo pattern ov er sev eral milliseconds corresponding to the entire portion ol the passenger compartment v olume ot interest is analv zed from multiple transducers and sometimes combined w ith the output from other transducers providing distance information to points on the items occupy ing the passenger compartment
- pattern recognition is central to the instant invention In the abov e-cited prior art except in that assigned to the current assignee ot the instant inv ention, pattern recognition which is based on training as exemplified through the use of neural networks, is not mentioned foi use in monitoring the interior passenger compartment or exterior env ironments ot the v ehicle Thus, the methods used to adapt such sv stems to a v ehicle are also not mentioned
- w generally mean any sy stem w hich processes a signal that is generated by an ob
- ect e g . representativ e ot a pattern of returned or receiv ed impulses, wav es or other physical property specific to and/or characteristic of and/or representativ e ot that obiect
- the signals processed are generall a series of electrical signals coming from transducers that are sensitive to acoustic ( ultrasonic ) or electromagnetic radiation (e.g.. visible light, infrared radiation, capacitance or electric and magnetic fields), although other sources of information are frequently included.
- Pattern recognition sy stems generally involv e the creation of the set of rules that permit the pattern to be recognized. These rules can be created by fuzzy logic sy stems, statistical correlations, or through sensor fusion methodologies as well as by trained pattern recognition systems such as neural networks.
- a trainable or a trained pattern recognition sy stem as used herein generally means a pattern recognition system which is taught to recognize various patterns constituted within the signals by subjecting the system to a v ariety of examples.
- test data is first obtained which constitutes a plurality of sets of returned waves, or wave patterns, from an object (or from the space in which the object w ill be situated in the passenger compartment, i.e.. the space abov e the seat ) and an indication of the identi fy of that object, (e.g.. a number of different objects are tested to obtain the unique wave patterns from each object).
- the algorithm is generated, and stored in a computer processor, and which can later be applied to provide the identity of an object based on the wav e pattern being received during use by a receiver connected to the processor and other information.
- an object sometimes applies to not only the object itself but also to its location and/or orientation in the passenger compartment.
- a rear facing child seat is a different object than a forward facing child seat and an out-of-position adult is a different object than a normally seated adult.
- the class may be one containing, for example, all rear facing child seats, one containing all human occupants, or all human occupants not sitting in a rear facing child seat depending on the purpose of the sy stem.
- the set or class will contain only a single element, i.e.. the person to be recognized.
- An "object " in a v ehicle or an "occupant " or “occupying item " of a seat may be a liv ing occupant such as a human or a dog. another liv ing organism such as a plant, or an inanimate object such as a box or bag of groceries or an empty child seat.
- Out-of-position as used for an occupant will generally means that the occupant, either the driver or a passenger, is sufficiently close to the occupant protection apparatus (airbag) prior to deploy ment that he or she is likely to be more seriously injured by the deployment ev ent itself than by the accident. It may also mean that the occupant is not positioned appropriately in order to attain beneficial, restraining effects of the deploy ment of the airbag. As for the occupant being too close to the airbag. this ty pically occurs when the occupant ' s head or chest is closer than some distance such as about 5 inches from the deploy ment door of the airbag module.
- transducer w ill mean only a receiv er
- Transducers include, tor example, capacitiv e. inductiv e, ultrasonic. electromagnetic (antenna. CCD. CMOS array s), weight measuring or sensing dev ices
- Adaptation represents the method by which a particular occupant sensing sy stem is designed and arranged for a particular v ehicle model It includes such things as the process bv which the number kind and location of v arious transducers is determined Tor pattern recognition systems, it includes the process by which the pattern recognition system is taught to recognize the desired patterns In this connection, it w ill usually include ( 1 ) the method of training. (2) the makeup ot the databases used for training testing and v alidating the particular system, or. in the case ot a neural network, the particular network architecture chosen.
- adaptation includes all ot the steps that are undertaken to adapt transducers and other sources of information to a particular v ehicle to create the sy stem w hich accurately identifies and determines the location of an occupant or other ob
- ect or v ehicle approaching another w ill generally mean the relativ e motion of the obiect toward the ehicle hav ing the anticipatory sensor system
- the coordinate sy stem used in general will be a coordinate sy stem residing in the target v ehicle
- the 'target" vehicle is the v ehicle which is being impacted
- This convention permits a general description to cover all ot the cases such as where ( i) a mov ing v ehicle impacts into the side ot a stationary v ehicle ( ⁇ ) where both vehicles are mov ing when they impact, or (in) where a vehicle is moving sideways into
- neural networks or neural fuzzy sy stems, as the pattern recognition technology and the methods ot adapting this to a particular vehicle, such as the training methods is important to this inv ention since it makes the monitoring system robust, reliable and accurate 1 he resulting algorithm created by the neural network program is usually only a tew hundred lines ot code written in the C computer language and is in general fewer lines than when the techniques of U S Pat Nos 5 008.946 ( Ando ).
- the resulting sy stems are easy to implement at a low cost making them practical for automotiv e applications
- the cost ol the ultrasonic transducers for example, is expected to be less than about S I in quantities ot one million per year
- the implementation ot the techniques ot the abo e-referenced patents requires expensiv e microprocessors while the implementation w ith neural networks and similar trainable pattern recognition technologies permits the use of low cost microprocessors typically costing less than about $5 in quantities of one million per year
- the present inv ention uses sophisticated software that dev elops trainable pattern recognition algorithms such as neural networks Usually the data is preprocessed.
- Both laser and non-laser optical systems in general are good at determining the location of objects w ithm the two dimensional plane of the image and a pulsed laser radar sy stem in the scanning mode can determine the distance of each part of the image from the receiver by measuring the time of flight through range gating techniques It is also possible to determine distance with the non-laser system by focusing, or stereographically if two spaced apart receiv ers are used and. in some cases, the mere location in the field of v iew can be used to estimate the position relativ e to the airbag. for example Finally, a recently dev eloped pulsed quantum well diode laser also prov ides inexpensiv e distance measurements as discussed in U S provisional patent application Serial No 60/1 14.507. filed December 3 1 1998
- a sy stem using these ideas is an optical sy stem which floods the passenger seat with infrared light coupled w ith a lens and a receiv er arrav e g CCD or C MOS array , which receiv es and display s the reflected light and an analog to digital converter ( ADC) w hich digitizes the output ot the CCD or CMOS and feeds it to an Artificial Neural Network ( ANN ) or other pattern recognition sy stem for analy sis 1 his sy stem uses an ultrasonic ti ansmitter and receiv er tor measuring the distances to the objects located in the passenger seat The receiv ing transducer feeds its data into an ADC and from there the converted data is directed into the ANN The same ANN can be used for both sy stems thereby providing full three-dimensional data tor the ANN to analyze This sv stem.
- ADC analog to digital converter
- the optical part can determine the location of the driver's ears, for example, and the phased array can direct a narrow beam to the location and determine the distance to the occupant's ears
- the speed ot sound limits the rate at which the position ot the occupant can be updated to approximately 10 milliseconds, w hich though sufficient for most cases, is marginal if the position of the occupant is to be tracked during a v ehicle crash
- ultrasound wav es are diffracted by changes in air density that can occur when the heater or air conditioner is operated or when there is a high-speed flow of air past the transducer
- the resolution ot ultrasound is limited by its wavelength and by the transducers, which are high Q tuned dev ices Typically , the resolution of ultrasound is on the order of about 2 to 3 inches
- the fields from ultrasonic transducers are difficult to control so that reflections from unwanted objects or surfaces add noise to the data Ultrasonics alone can be used in several configurations tor monitoring the interior of a passenger compartment of an automobile as described in the abov e-
- the mam focus of the instant invention is the process and resulting apparatus of adapting the sy stem in the patents and patent applications referenced abov e for the detection of the presence of an occupied child seat in the rear facing position or an out-of-position occupant and the detection ot an occupant in a normal seating position
- the system is designed so that in the former two cases, deploy ment of the occupant protection apparatus (airbag) may be controlled and possibly suppressed and in the latter, it will be controlled and enabled
- ANN Artificial Neural Network
- the pattern is obtained from four ultrasonic transducers that cov er the front passenger seating area This pattern consists of the ultrasonic echoes bouncing off of the objects in the passenger seat area
- the signal from each of the tour transducers consists ol the electrical image ot the return echoes, which is processed by the electronics T he electronic processing comp ⁇ ses amplification logarithmic compression rectification and demodulation ( band pass filtering), followed by discretization ( sampling) and digitization of the signal
- the only software processing required, before this signal can be fed into the artificial neural network is normalization ( i e . mapping the input to numbers between 0 and 1 ) Although this is a fair amount of processing, the resulting signal is still considered
- This invention is a svstem designed to identify , locate and monitor occupants, including their parts, and other objects in the passenger compartment and in particulai an occupied chi ld seat in the rear facing position or an out-of-position occupant, by illuminating the contents of the v ehicle ith ultrasonic or electromagnetic radiation, for example, by transmitting radiation w aves from a wav e generating apparatus into a space above the seat, and receiv ing radiation modified bv passing through the space abov e the seat using two or more transducers properly located in the v ehicle passenger compartment, in specific predetermined optimum location More particularly , this inv ention relates to a sy stem including a plurality of transducers appropriately located and mounted and which analv ze the receiv ed radiation from any object which modifies the waves, in order to achieve an accuracy of recognition heretofore not possible Outputs from the receiv ers are analyzed by appropriate computational means
- the information obtained bv the identification and monitoring system is used to affect the operation of some other system, component or dev ice in the v ehicle and particularly the passenger and/or driv er airbag sy stems airbag w hich may include a front airbag. a side airbag. a knee bolster, or combinations of the same
- the information obtained can be used for controlling or affecting the operation of a multitude ot other v ehicle sy stems
- the v ehicle interior monitoring sy stem in accordance w ith the inv ention is installed in the passenger compartment of an automotiv e vehicle equipped with a occupant protection apparatus, such as an inflatable airbag. and the v ehicle is subjected to a crash of sufficient sev erity that the crash sensor has determined that the protection apparatus is to be deployed the sy stem has determined ( usually prior to the deploy ment) w hether a child placed in the rear facing position in the child seat is present and if so a signal has been sent to the control circuitry that the airbag should be controlled and most likclv disabled and not deployed in the crash It must be understood though that instead ot suppressing deploy ment it is possible that the deploy ment may be controlled so that it might prov ide some meaningful protection for the occupied rear-tacmg child seat
- the sy stem dev eloped using the teachings ol this inv ention also determines the position of the v eh
- a first embodiment of a method of developing a system for determining the occupancy state of a seat in a passenger compartment of a vehicle comprises the steps of mounting transducers in the vehicle, which transducers would be affected by the occupancy state of the seat, forming at least one database comprising multiple data sets, each data set representing a different occupancy state of the seat and being formed by receiving data from the transducers while the seat is in that occupancy state, and processing the data received from the transducers, and creating a first algorithm from the database(s) capable of producing an output indicative of the occupancy state of the seat upon inputting a new data set representing an occupancy state of the seat.
- the new data set would be formed, e.g.. during use of the vehicle after the algorithm is installed in the control circuitry of the vehicle.
- the first algorithm may be created by inputting the database(s) into an algorithm-generating program, and running the algorithm- generating program to produce the first algorithm.
- the first algorithm could be a neural network algorithm. in which case, the back propagation method could be used when generating the neural network algorithm.
- the occupancy states of the seat include occupancy of the seat by an object selected from the group comprising occupied and unoccupied rear facing infant seats, forward facing humans, out-of-position humans, occupied and unoccupied forward facing child seats and empty seats.
- the occupancy states of the seat should also include occupancy by the objects in multiple orientations and/or having at least one accessory selected from a non-exclusive group comprising newspapers, books, maps, bottles, toys, hats, coats, boxes, bags and blankets.
- the data can be pre-processed prior to being formed into the data sets. This may entail using data created from features of the data in the data set. which features might be selected from a group comprising the normalization factor, the number of data points prior to a peak, the total number of peaks, and the mean or variance of the data set. Also, the data sets could be mathematically transformed using normalization, truncation, logarithmic transformation, sigmoid transformation, thresholding, averaging the data over time. Fourier transforms and/or wav elet transforms 1 urther. pre-processing could entail subtracting data in one data set from the corresponding data in another data set to create a third data set ot differential data
- the processing step may comprise the step of conv erting the analog data from the transducers to digital data and combining the digital data from a plurality ot the transducers to form a v ector comprising a string of data from each ot the transducers
- the first algorithm is created such that upon inputting a v ector from a new data set w ill produce an output representing the occupancy state of the v ehicle seat
- the vectors in the database can be normalized so that all v alues ot the data that comprise each v ector are between a maximum and a minimum
- Another method of dev eloping a sy stem tor determining the occupancy state of the v ehicle seat in the passenger compartment ot a v ehicle comprises the steps of forming data sets by obtaining data representativ e ot v arious occupy ing objects at various positions in the passenger compartment and operating on at least a portion ot the data to reduce the magnitude ot the largest data values in a data set relative to the smallest data v alues, terming a database comprising multiple data sets, and creating an algorithm from the database capable ot producing an output indicative ot the occupancy state of the vehicle seat upon inputting a data set representing an occupancy state ot the seat
- Operating on the data mav entail using an approximate logarithmic transtormation lunction
- a method ot dev eloping a database for use in dev eloping a system for determining the occupancy state of a v ehicle seat in accordance with the inv ention comprises the steps of mounting transducers in the v ehicle and which would be affected by the occupancy state of the seat, providing the seat with an initial occupancy state, receiv g data from the transducers, processing the data from the transducers to form a data set representative of the initial occupancy state of the v ehicle seat, changing the occupancy state of the seat and repeating the data collection process to form another data set.
- Another method of dev eloping a system for determining the occupancy state of a passenger compartment seat of a v ehicle comprises the steps of mounting a plurality of ultrasonic transducers in the v ehicle (which transducers would be affected by the occupancy state of the seat), receiv ing an analog signal from each of the transducers, processing the analog signals from the
- said data processing comprising the steps ot demodulation sampling and digitizing of the transducer data to create a data set of digital data, forming a database comprising multiple data sets and creating at least one algorithm from the database capable ot producing an output indicativ e ol the occupancy state of the seat upon inputting a new data set representing an occupancy state of the seat
- Still another method ot developing a sv stem for determining the occupancy state of a vehicle seat in a passenger compartment comprises the steps ot mounting a set of transducers on the vehicle, receiv ing data from the transducers, processing the data from transducers to form a data set representativ e of the occupancy state of the v ehicle, forming a database comprising multiple data sets. creating an algorithm from the database capable ot producing an output indicative of the occupancy state of the vehicle seat upon inputting a new data set and developing a measure ol system accuracy At least one transducer is remov ed from the transducer set.
- a new database is created containing data only trom the reduced number ot transducers, a new algorithm is dev eloped based on the new database and tested to determine the new sy stem accuracy
- the process ot remov ing transducers, algo ⁇ tnm development and testing is continued until the minimum number of sensors is determined which produces an algorithm having desired accuracy
- the transducers are selected trom a group consisting ot ultrasonic transducers. optical sensors capacitiv e sensors, weight sensors, seat position sensors, seatback position sensors, seat belt buckle sensors, seatbelt payout sensors infrared sensors, inductiv e sensors and radar sensors
- Yet another method of developing a sy stem for determining the occupancy state of the driv er and passenger seats of a v ehicle comprises the steps of mounting ultrasonic transducers hav ing different transmitting and receiv ing frequencies in a v ehicle such that transducers hav ing adiacent frequencies are not withm the direct ultrasonic field of each other, receiv ing data trom the transducers processing the data from the transducers to form a data set representativ e ot the oceupancv state of the v ehicle, forming at least one database comprising multiple data sets and creating at least one algorithm from the at least one database capable of producing an output indicative ot the occupancy state ol a vehicle seat upon inputting a new data set
- FIG 1 shows a seated-state detecting unit dev eloped in accordance w ith the present inv ention and the connections between ultrasonic or electromagnetic sensors, a weight sensoi . a reclining angle detecting sensor, a seat track position detecting sensor, a heartbeat sensor, a motion sensor a neural network circuit. and an airbag sy stem installed within a v ehicle compartment
- TIG. 2 is a perspectiv e v iew ot a vehicle containing two adult occupants on the front seat with the vehicle shown in phantom illustrating one preferred location of the ultrasonic transducers placed according to the methods taught in this invention
- FIG 3 is a v iew as in F IG 2 w ith the passenger occupant replaced by a child in a forward facing child seat
- FIG 4 is a v iew as in 1 IG 2 w ith the passenger occupant replaced bv a child in a rearwar ⁇ facing child seat
- FIG 5 is a v lew as in FIG 2 with the passenger occupant replaced by an infant in an infant seat
- F IG 6 is a diagram illustrating the interaction ot two ultrasonic sensors and how this interaction is used to locate a circle is space
- FIG 7 is a v iew as in FIG 2 with the occupants remov ed illustrating the location of two circles in space and how they intersect the v olumes characteristic ot a rear facing child seat and a larger occupant
- FIG 8 illustrates a preferred mounting location ot a three-transducer sv stem
- FIG 9 illustrates a preferred mounting location ot a four-transducer sv stem
- FIG 10 is a plot show ing the target v olume discrimination tor two transducers
- FIG 1 1 illustrates a preferred mounting location ot a eight-transducer sy stem
- FIGS 12- 19 are setups used for training ot a neural network in accordance w ith the inv ention
- FIG 20 is a chart ot four t pical raw signals which are combined to constitute a v ector
- Sy stem Adaptation inv olv e the process by which the hardware configuration and the software algorithms are determined tor a particular v ehicle Each v ehicle model or plattorm will most likely hav e a different hardware configuration and different algorithms Some ot the v rious aspects that make up this process are as follo s
- the automatic recording ot tne vehicle setup including seat, seat back, headrest, window, v isor. armrest positions to help insure data integrity
- the normalization factor The total number ot peaks The vector mean or v a ⁇ ance
- the process of adapting the sv stem to the v ehicle begins with a survey of the v ehicle model Any existing sensors, such as seat position sensors, seat back sensors, etc are immediate candidates for inclusion into the system Input trom the customer w ill determine what ty pes ot sensors would be acceptable for the final system
- These sensors can include seat structure mounted weight sensors, pad type weight sensors, pressure ty pe weight sensors seat fore and aft position sensors, seat v ertical position sensors, seat angular position sensors seat back position sensors, headrest position sensors, ultrasonic occupant sensors, optical occupant sensors, capacitive sensors, inductiv e sensors, radar sensors, v ehicle velocity and acceleration sensors, brake pressure, seatbelt force, pay out and buckle sensors etc
- a candidate array of sensors is then chosen and mounted onto the v ehicle
- the v ehicle is also instrumented so that data input by humans is minimized
- a standard set of vehicle setups is chosen for initial trial data collection purposes Ty pically, the initial trial w ill consist of between 20.000 and 100.000 setups
- the trial database w ill also include env ironmental effects such as thermal gradients caused by heat lamps and the operation ot the air conditioner and heatei
- env ironmental effects such as thermal gradients caused by heat lamps and the operation ot the air conditioner and heatei
- a sample of such a matrix is presented in Appendix 1
- the trial database ill be scanned for v ectors that v ield erroneous results (which would hkelv be considered bad data)
- a study ot those v ectors along with vectors from associated in time cases are compared with the photographs to determine whether there is erroneous data present I f so an attempt is made to determine the cause of the erroneous data If the cause can be found for example if a v oltage spike on the power line corrupted the data then the vector will be removed from the database and an attempt is made to correct the data collection process so as to remove such disturbances
- the training database This will be the largest database initiall y collected and w ill cover such setups as listed for example in Appendix I
- the training database which mav contain 500 000 or more v ectors will be used to begin training of the neural network While this is taking place additional data will be collected according to Appendix 1 of the independent and validation databases
- the training database has been selected so that it uniformly cov ers all seated states that are known to be hkelv to occur in the v ehicle
- the independent database mav be similar in makeup to the training database or it mav evolv e to more closelv conform to the occupancy state distribution of the validation database
- the independent database is used to check the accuracy ot the neural network and to reject a candidate neural network design if its accuracy measured against the independent database is less than that ot a prev ious network architecture
- the independent database is not actually used in the training of the neural network nevertheless it has been found that it significantly influences the network structure Therefore a third database the v alidation or real world database is used as a final accuracy check of the chosen sv stem It is the accuracy against this v alidation database that is considered to be the sv stem accuracy
- the v alidation database is composed of v ectors taken trom setups which closelv correlate with v ehicle occupancy in real cars on the roadwav Initially the training database is the largest of the three databases As time and resources permit the independent database w hich perhaps starts out w ith 100 000 vectors will continue to grow until it becomes approximately the same size as the training database T he validation database on the other hand w ill tv picaliv start out with as tew as 50 000 vectors Howev er as the hardware configuration is frozen the validation database will continuously grow until in some cases it actuall y becomes larger than the training database This is because near the end ot the
- a series of networks would be trained using all combinations of six sensors from the 20 available.
- the activ ity would require a prohibitively long time.
- Certain constraints can be factored into the sy stem from the beginning to start the pruning process. For example, it would probabl y not make sense to have both optical and ultrasonic sensors present in the same system since it would complicate the electronics. In fact, the automobile manufacturer may hav e decided initially that an optical sy stem would be too expensive and therefore would not be considered.
- the inclusion of optical sensors therefore, serves as a way of determining the loss in accuracy as a function of cost.
- Various constraints therefore, usually allow the immediate elimination of a significant number of the initial group of sensors. This elimination and the training on the remaining sensors provides the resulting accuracy loss that results.
- the next step is to remove each of the sensors one at a time and determine which sensor has the least effect on the system accuracy. This process is then repeated until the total number of sensors has been pruned down to the number desired by the customer. At this point, the process is reversed to add in one at a time those sensors that were removed at previous stages. It has been found, for example, that a sensor that appears to be unimportant during the early pruning process can become very important later on. Such a sensor may add a small amount of information due to the presence of various other sensors. Whereas the v arious other sensors, however, may yield less information than still other sensors and. therefore may hav e been remov ed during the pruning process.
- the automobile manufacturer may desire to have the total of 6 transducers in the final sy stem, however, when shown the fact that the addition of one or two additional sensors substantially increases the accuracy of the system, the manufacturer may change his mind. Similarly , the initial number of sensors selected may be 6 but the analy sis could show that 4 sensors give substantially the same accuracy as 6 and therefore the other 2 can be eliminated at a cost saving.
- the v ehicle While the pruning process is occurring, the v ehicle is subjected to a v ariety ot road tests and would be subjected to presentations to the customer
- the road tests are tests that are run at different locations than where the fundamental training took place It has been lound that unexpected environmental factors can influence the performance ot the sy stem and therefore these tests can provide critical information I he sy stem, therefore, w hich is installed in the test v ehicle should hav e the capability ot recording sy stem failures
- This recording includes the output of all of the sensors on the v ehicle as well as a photograph ot the v ehicle setup that caused the error
- This data is later analyzed to determine whether the training, independent or v alidation setups need to be modified and/or whether the sensors or positions of the sensors require modification
- the v ehicle is again subjected to real world testing on highways and at customer demonstrations
- the system described so tar has been based on the use ot a single neural network It is frequently necessary to use multiple neural networks or other pattern recognition sy stems
- a vehicle seat there are really two requirements The first requirement is to establish what is occupy ing the seat and the second requirement is to establish where that object is located Generally a great deal of time, typically many seconds, is av ailable for determining whether a forward facing human or an occupied or unoccupied rear facing child seat, for example, occupies the vehicle seat.
- the data that is fed to the pattern recognition sy stem typically will usually not be the raw v ectors of data as captured and digitized from the v arious transducers Ty pical lv . a substantial amount ot preprocessing of the data is undertaken to extract the important information trom the data that is fed to the neural network This is especially true in optical sy stems and where the quantity ot data obtained, if all were used by the neural network, would require v er, expensiv e processors
- the output from the pattern recognition sv stem is usually based on a snapshot of the output of the v arious transducers Thus, it represents one epoch or time period.
- the accuracy ot such a decision can usually be substantially improv ed if prev ious decisions from the pattern recognition sy stem are also considered
- the results of many decisions are av eraged together and the resulting av eraged decision is chosen as the correct decision
- the position ot the occupant must be known at that particular epoch and cannot be av eraged with his prev ious position
- there is information in the prev ious positions that can be used to improv e the accuracy of the current decision For example, if the new decision says that the occupant has moved six inches since the prev ious decision, and.
- an occupancy position v ersus time curv e can be fitted using a v ariety of techniques such as the least squares regression method, to the data trom prev ious 10 epochs, for example This same type of analy sis could also be applied to the v ector itself rather than to the final decision thereby correcting the data prior to its being entered into the pattern recognition system
- a pattern recognition system such as a neural network
- can sometimes make totally irrational decisions This typically happens when the pattern recognition sv stem is presented with a data set or vector that is unlike any vector that has been in its training set
- the v ariety of seating states ot a vehicle is unlimited Ev ery attempt is made to select from that unlimited universe a set of representative cases Nevertheless, there will alway s be cases that are significantly different from anv that hav e been previously presented to the neural network
- the final step therefore, to adapting a sy stem to a v ehicle, is to add a measure of human intelligence Sometimes this goes under the heading of fuzzy logic and the resulting system has been termed in some cases a neural fuzzy sy stem In some cases, this takes the form of an observ er studv mg failures ot the sv stem and coming up w ith rules and that say .
- FIG 1 shows a passenger seat 1 to which an adjustment apparatus including a seated-state detecting sv stem developed according to the present inv ention may be applied
- the seat 1 includes a horizontall y situated bottom seat portion 2 and a vertically oriented back portion 3
- the seat portion 2 is prov ided w ith weight measuring means i t one or more weight sensors 6 and 7 that determine the weight ot the obiect occupy ing the seat it anv
- the coupled portion between the seated portion 2 and the back portion 3 (also referred to as the seatback) is provided with a reclining angle detecting sensor 9 which detects the tilted angle ot the back portion 3 relative to the seat portion 2
- the seat portion 2 is prov ided with a seat track position-detecting sensor 10
- the seat track position detecting sensor 10 fulfills a role of detecting the quantity of movement of the seat 1 which is moved trom a back reference position indicated by the dotted chain line Emb
- the weight measuring means such as the sensors 6 and 7 are associated with the seat and can be mounted into or below the seat portion 2 or on the seat structure for example for measuring the weight applied onto the seat
- the weight mav be zero it no occupy ing item is present
- Sensors 6 and 7 may represent a plurality of different sensors which measure the weight applied onto the seat at different portions thereof or for redundancy purposes for example such as bv means of an airbag or bladder • > in the seat portion 2 T he bladder 3 mav hav e one or more compartments
- Such sensors mav be in the form of strain force or pressure sensors which measure the force or pressure on the seat portion 2 or seat back 3 displacement measuring sensors which measure the displacement of the seat surface or the entire seat 1 such as through the use ot strain gages mounted on the seat structural members such as 7 or other appropriate locations, or sv stems which conv ert displacement into a pressure w herein a pressure sensor can be used as a measure of weight
- An ultrasonic or optical sensor sv stem 12 is mounted on the upper portion ot the front pillar.
- A- Pillar. ot the v ehicle and a similar sensor sy stem 1 1 is mounted on the upper portion ot the intermediate pillar B-Pillar
- the outputs ot the transducers I I and 12 are input to a band pass filter 20 through a multiplex circuit 19 w hich is sw itched in sy nchronization w ith a timing signal from the ultrasonic sensor driv e circuit 18 and then is amplified by an amplifier 21
- the band pass filter 20 removes a low frequency wave component trom the output signal and also remov es some of the noise
- the env elope wave signal is input to an analog/digital conv erter (ADC) 22 and digitized as measured data
- ADC analog/digital conv erter
- Each of the measured data is input to a normalization circuit 24 and normalized
- the normalized measured data is input to the neural network (circuit) 25 as wav e data
- the output of the weight sensor(s) 6 and 7 is amplified bv an amplifier 26 coupled to the weight sensor( s) 6 and 7 and the amplified output is input to an analog digital conv erter and then directed to the neural network 25 of the processor means
- the reclining angle detecting sensor 9 and the seat track position-detecting sensor 10 are connected to appropriate electronic circuits
- a constant-current can be supplied from a constant-current circuit to the reclining angle detecting sensor 9 and the reclining angle detecting sensor 9 conv erts a change in the resistance value on the tilt ot the back portion 3 to a specific voltage
- This output voltage is input to an analog/digital converter 28 as angle data, I e representativ e ot the angle between the back portion 3 and the seat portion 2
- a constant current can be supplied from a constant-current circuit to the seat track position detecting sensor 10 and the seat track position detecting sensor 10 converts a change in the resistance v alue based on the track position of the seat portion 2 to a specific voltage
- This output v oltage is input to an analog/digital converter 29 as seat track data
- the outputs of the reclining angle-detecting sensor 9 and the seat track position-detecting sensor 10 are input to the analog digital converters (ADC) 28 and 29 respectiv
- the neural network 25 is directly connected to the ADCs 28 and 29 the ADC associated with amplifier 25 and the normalization circuit 24 As such information from each of the sensors in the sv stem (a stream ot data) is passed directly to the neural network 25 for processing thereby
- the streams ot data from the sensors are not combined prior to the neural network 25 and the neural network is designed to accept the separate streams ot data ( e g .
- the neural network 25 thus includes or incorporates an algorithm de ⁇ v ed bv training in the manners discussed abov e and below
- v ehicular components or systems such as the airbag sy stem in consideration of the current occupanc y state ot the seat
- a section ot the passenger compartment of an automobile is shown generally as 100 in FIG 2
- a driver 101 ot a v ehicle sits on a seat 102 behind a steering wheel not shown, and an adult passenger 103 sits on seat 104 on the passenger side
- transducers are positioned in the passenger compartment 100.
- one transducer 1 10 is arranged on the headliner adjacent or in proximity to the dome light and the other transducer 1 1 1 is arranged on the center of the top of the dashboard or instrument panel of the v ehicle Fhe methodology leading to the placement of these transducers is central to the instant inv ention as explained in detail below
- the sy stem dev eloped in accordance with this inv ention will reliably detect that an occupant is sitting on seat 104 and deployment ot the airbag is enabled in the ev ent that the vehicle experiences a crash
- Transducers 1 10 1 1 1 are placed with their separation axis parallel to the separation axis of the head, shoulder and rear facing child seat volumes of occupants of an automotiv e passenger seat and in v iew of this specific positioning, are capable of distinguishing the different configurations In addition to the ultrasonic transducers 1 10 1 1 1.
- weight-measuring sensors 210. 21 1. 212. 214 and 215 are also present These weight sensors may be ot a variety of technologies including as illustrated here, strain-measuring transducers attached to the v ehicle seat support structure as described in more detail in co-pending U S patent application Serial No 08/920.822 Naturally other weight sv stems can be utilized including sv stems that measure the deflection of or pressure on the seat cushion The weight sensors described here are meant to be illustrativ e of the general class of weight sensors and not an exhaustive list of methods of measuring occupant weight In FIG 3.
- a forward facing child seat 120 containing a child 121 replaces the adult passenger 103 as shown in FIG 2
- the airbag it is usually required that the airbag not be disabled in the event of an accident Howev er, in the ev ent that the same child seat is placed in the rearward facing position as shown in FIG 4. then the airbag is usually required to be disabled since deployment of the airbag in a crash can seriously iniure or even kill the child
- the airbag should be disabled for the reasons discussed abov e
- the deployment could be controlled to prov ide protection for the child e g .
- the disabling or enabling of the passenger airbag relativ e to the item on the passenger seat may in tact be desirable to disable the airbag and in other cases to deploy a depowered airbag
- the selection of when to disable, depovver or enable the airbag. as a function of the item in the passenger seat and its location is made during the programming or training stage ot the sensor system and.
- the criteria set forth abov e will be applicable I e enabling airbag deployment for a forward facing child seat and an adult in a proper seating position and disabling airbag deplovment for a rearward facing child seat and infant and tor anv occupant who is out-of-position and in close proximity to the airbag module
- the sensor system dev eloped in accordance w ith the inv ention mav howev er be programmed according to other criteria
- Some of these sy stems appear to work as long as the child seat is properly placed on the seat and belted in So called lag sy stems "for example w herebv a dev ice is placed on the child seat which is electromagneticallv sensed by sensors placed within the seat hav e not prov en reliable by themselves but can add information to the ov erall sv stem When used alone, thev lunction well as long as the child seat is restrained bv a seatbelt.
- the object mav be at a point on the surface of a three-dimensional spherical segment having its origin at the transducer and a radius equal to the distance
- two transducers such as 1 10 and 1 1 1 m FIG 6.
- both transducers receive a reflection from the same object, which is facilitated by proper placement of the transducers the timing of the reflections depends on the distance from the ob
- each transducer knows that the ob
- the airbag would not be disabled since its deployment in a crash is desired On the other hand, if a circle is at a location occupied only by a rear facing child seat, the airbag would be disabled
- transducer B is hkelv to pickup the rear ot the occupant s head and transducer ⁇ the front This makes the situation more difficult toi an engineer looking at the data to analy ze It has been tound that pattern recognition technologies are able to extract the information trom these situations and through a proper application of these technologies, an algorithm can be dev eloped, w hich when installed as part of the system for a particular vehicle, the system accurately and reliably differentiates between a forward facing and rear facing child seat, for example, or an in-position or out-of-position forward facing human being
- a method of transducer location which provides unique information to differentiate between d ) a forward facing child seat or a forward properly positioned occupant where airbag deployment is desired and (n) a rearward facing child seat and an out-of-position occupant w here airbag deployment is not desired ln actuality , the algorithm used to implement this theory does not directly calculate the surface ot spheres or the circles of interaction ot spheres Instead, a pattern recognition sv stem is used to differentiate airbag-deploy ment desired cases from those w here the airbag should not be deploy ed For the pattern recognition sy stem to accurately perform its function, howev er, the patterns presented to the system must hav e the requisite information That is.
- a pattern ot reflected waves trom an occupying item in a passenger compartment to v arious transducers must be uniquely different for cases where airbag deplovment is desired trom cases where deploy ment is not desired.
- the theory described above and in more detail below teaches how to locate transducers w ithin the vehicle passenger compartment so that the patterns ot reflected wav es w ill be easily distinguishable for cases where airbag deplovment is desired trom those where deployment is not desired In the case presented thus far.
- the use ot onlv two transducers can result in the desired pattern differentiation when the v ehicle geometry is such that two transducers can be placed such that the circles D (airbag enabled) and E (airbag disabled ) tall outside of the transducer tield cones except w here thev are in the critical regions w here positiv e identification of the condition occurs
- the aiming and field angle of the transducers are important factors to determine in adapting a sy stem to a particular vehicle
- Transducer 1 13 is positioned on the ceiling of the v ehicle close to the passenger side door
- lines connecting the transducers C and D and the transducers A and B are substantially parallel permitting an accurate determination ot asy mmetry and thereby object rotation
- FIG 5 there is the infant only seat as shown in FIG 5 which is for occupants weighing up to about 20 pounds This is designed to be only placed in the rear facing position
- the second which is illustrated in FIGS 2 and 3 is for children trom about 20 to about 40 pounds and can be used in both the forward and rear facing position and the third is tor use only in the forward facing position and is for children weighing over about 40 pounds All of these seats as well as the unique models are used in test setups according to this inv ention for adapting a sy stem to a v ehicle
- wide v ariations are used tor the occupants including size, clothing and activities such as reading maps or newspapers. leaning forward to adjust the radio, for example Also included are cases where the occupant puts his/her feet on the dashboard or otherwise assumes a wide v ariety of unusual positions
- the total number ot configurations which are used to tram the pattern recognition sy stem can exceed 500.000
- the goal is to include in the configuration training set representations of all occupancy states that occur in actual use Since the system is highly accurate making the correct decision for cases which are similar to those in the training set. the total sy stem accuracy increases as the size of the training set increases providing the cases are all distinct and not copies of other cases
- thermal gradients or thermal instabilities are particularly important since sound wav es can be significantly diffracted by density changes in air
- thermal gradients or instability there are two aspects of the use of thermal gradients or instability in training
- a platform is an automobile manufacturer s designation of a group of vehicle models that are built on the same v ehicle structure
- Occupant size position (zones), clothing etc Child seat type size position etc
- step 13 inv olv e the use ot transducers with full horns mounted on the surfaces ot the interior passenger compartment At some point, the actual transducers which are to be used in the final v ehicle must be substituted for the trial transducers This is either done prior to step 13 or at this step
- This process inv olv es designing transducer holders that blend with the v isual surfaces ot the passenger compartment so that they can be cov ered with a properly designed grill that helps control the field and also serv es to retain the esthetic quality of the interior This is usually a lengthy process and inv olv es sev eral consultations with the customer
- the system is trained and tested w ith situations representativ e ot the manufacturing and installation tolerances that occur during the production and deliv ery of the v ehicle as well as usage and deterioration effects
- the system is tested with the transducer mounting positions shifted by up to one inch in any direction and rotated by up to 15 degrees, with a simulated accumulation of dirt and other v ariations
- This tolerance to vehicle v ariation also sometimes permits the installation of the system onto a different but similar model vehicle with, in many cases, only minimal retraining ot the system
- the speed of sound v aries with temperature, humidity , and pressure This can be compensated tor by using the tact that the geometry between the transducers is known and the speed of sound can therefore be measured
- the speed of sound can be measured by one transducer, such as transducer 1 10 in FIG 5.
- the sy stem operates w ith same accuracy regardless ot the temperature, humidity or atmospheric pressure It mav ev en be possible to use this technique to also automatically compensate tor any effects due to wind v elocity through an open window An additional benefit of this sy stem is that it can be used to determine the v ehicle interior temperature for use by other control systems within the vehicle since the v ariation in the v elocity ot sound is a strong tunction of temperature and a weak function of pressure and humidity
- the electronic control module that is part of the sv stem is located in generally the same environment as the transducers, another method of determining the temperature is av ailable
- This method utilizes a device and whose temperature sensitivity is known and which is located in the same box as the electronic circuit
- an existing component on the printed circuit board can be monitored to giv e an indication of the temperature
- the diodes in the log comparison circuit have characteristics that their resistance changes in a known manner with temperature It can be expected that the electronic module will generally be at a higher temperature than the surrounding environment however, the temperature difference is a known and predictable amount Thus, a reasonably good estimation of the temperature in the passenger compartment can also be obtained in this manner
- Another important feature ot a svstem.
- the determination of the occupancy state can be substantially improv ed by using successiv e observ ations ov er a period of time This can eithei be accomplished by av eraging the data prior to insertion into a neural network, or alternately the decision of the neural network can be av eraged This is known as the categorization phase of the process During categorization the occupancy state of the vehicle is determined Is the vehicle occupied by the forward facing human, an empty seat, a rear facing child seat, or an out-of-position human 7 Typically many seconds ot data can be accumulated to make the categorization decision When a driv er senses an impending crash, on the other hand, he or she will ty pically slam on the brakes to trv to slow vehicle prior to impact I t an occupant is unbelted, he or she will begin moving toward the airbag during
- One method is to determine the location of the occupant using the neural network based on prev ious training The motion of the occupant can then be compared to a maximum likelihood position based on the position estimate of the occupant at previous v ectors
- the measured position of the occupant can be corrected based on his prev ious positions and known v elocity
- an accelerometer is present in the v ehicle and if the acceleration data is av ailable tor this calculation a much higher accuracy prediction can be made
- 500 data points per v ector mav be collected and ted to the neural network du ⁇ ng the training stage and. after careful pruning, the final number ot data points to be used by the v ehicle mounted sy stem may be reduced to 150.
- This technique of using the neural network algorithm-generating program to prune the input data is an important teaching of the present invention By this method, the advantages of higher resolution transducers can be optimally used without increasing the cost of the electronic v ehicle mounted circuits Also, once the neural network has determined the spacing of the data points, this can be fine-tuned, for example, by acquiring more data points at the edge of the keep out zone as compared to positions well into the safe zone.
- the initial technique is done be collecting the full 500 data points, for example, while in the system installed in the vehicle the data digitization spacing can be determined by hardware or software so that only the required data is acquired
- An alternate method of obtaining some of the adv antages of the parallel neural network architecture described abov e is to form a single neural network but where the nodes of one or more ot the hidden lay ers are not all connected to all of the input nodes.
- the second hidden layer is chosen, all ot the notes trom the previous hidden layer are not connected to all of the nodes of the subsequent layer
- the alternate groups of hidden lay er nodes can then feed to different output notes and the results ot the output nodes combined, either through a neural network training process into a single decision or a voting process This latter approach retains most of the advantages of the parallel neural network while substantially reducing the computational complexity
- modular neural networks improve the accuracy of the sy stem by div iding up the tasks For example, if a sy stem is to be designed to determine the ty pe of tree and the ty pe of animal in a particular scene, the modular approach would be to first determine whether the object of interest is an animal or a tree and then use separate neural networks to determine ty pe ot tree and the ty pe ot animal When a human looks at a tree he is not ask himself is that a tiger or a monkey Modular neural network sy stems are efficient since once the categorization decision is made, the seat is occupied by forward facing human, for example, the location of that obiect can be determined more accurately and without requiring increased computational resources
- the sy stem suffers only a slight reduction in accuracy ev en if two ot the transducers are covered so as to make them inoperable It is important in order to obtain the full adv antages ot the sy stem when a trans ⁇ ucer is blocked that the training and independent databases contains many examples of blocked transducers I f the patterr recognition system, the neural network in this case, has not been trained on a substantial number of blocked transducer cases, it w ill not do a good
- Other techniques which mav or mav not be part of the process of designing a svstem for a particular application include the following
- the back propagation neural network is a v erv successful general- purpose network Howev er for some applications there are other neural network architectures that can perform better If it has been found, for example, that a parallel network as described above results in a significant improvement in the system, then, it is likely that the particular neural network architecture chosen has not been successful in ret ⁇ ev mg all ot the information that is present in the data. In such a case an RCE. Stochastic. Logicon Projection, or one ot the other approximately 30 types ot neural network architectures can be tried to see it the results improv e This parallel network test, therefore, is a valuable tool for determining the degree to which the current neural network is capable of using efficiently the available data
- the training database must be flat meaning that all of the occupancy states that the neural network must recognize must be approximately equally represented in the training database Ty pically , the independent database has approximately the same makeup as the training database
- the validation database on the other hand, ty pically is represented in a non-flat basis with representative cases from real world experience Since there is no need for the validation database to be flat, it can include many of the extreme cases as well as being highly biased towards the most common cases This is the theory that is currently being used to determine the makeup of the various databases The success of this theory continues to be challenged by the addition of new cases to the validation database W hen significant failures are discov ered in the v alidation database, the training and independent databases are modified in an attempt to remov e the failure
- a neural network sy stem can frequently be aided if additional data is inputted into the network
- One example can be the number ot 0 data points before the first peak is experience Vternatelv . the exact distance to the first peak can be determined prior to the sampling ot the data
- Other features can include the number ot peaks the distance between the peaks, the width ot the largest peak the normalization factor, the v ector mean or standard dev iation, etc.
- v isual markings are placed such that a technician can observ e that the test occupant remains within the required zone tor the particular data taking exercise
- a laser diode is used to create a v isual line in the space that represents the boundary of the keep out zone or other desired zone boundary
- the adaptation process begins w ith a selection ot candidate transducers tor a particular v ehicle model This selection is based on such considerations as cost, alternate uses of the system other than occupant sensing, vehicle interior passenger compartment geometry , desired accuracy and reliability v ehicle aesthetics, v ehicle manufacturer preferences, and others
- a candidate set of transducers Once a candidate set of transducers has been chosen, these transducers are mounted in the test v ehicle according to the teachings of this inv ention
- the v ehicle is then subjected to an extensiv e data collection process wherein v arious objects are placed in the v ehicle at v arious locations as described below and an initial data set is collected ⁇ pattern recognition sv
- the design process begins with a surplus of sensors plus an objective as to how many sensors are to be in the final v ehicle installation
- the adaptation process can determine which of the transducers are most important and which are least important and the least important transducers can be eliminated to reduce sy stem cost and complexity
- the Artificial Neural Network that tor s the "brains " of the Occupant Spatial Sensor needs to be trained to recognize airbag enable and disable patterns
- the most important part of this training is the data that is collected in the v ehicle, w hich provides the patterns corresponding to these respective configurations
- Manipulation ot this data is appropriate if this enhances the information contained in the data important too.
- the ultimate test tor all methods and filters is their effect on the network performance against real world situations
- the Occupant Spatial Sensor uses an artificial neural network (ANN) to recognize patterns that it has been trained to identify as either airbag enable or airbag disable conditions
- the pattern is obtained from four ultrasonic transducers that cov er the front passenger seating area
- This pattern consists of the ultrasonic echoes from the objectss in the passenger seat area T
- he signal trom each of the tour transducers consists of the electrical image ot the return ecnoes w ich is processed bv the electronics
- the electronic processing compnses amplification ( logarithmic compression ) rectification, and demodulation (band pass filtering), followed by discretization ( sampling ) and digitization ot the signal
- the only software processing required, before this signal can be fed into the artificial neural network, is normalization ( l e mapping the input to numbers between 0 and 1 ) Although this is a fair amount of processing, the resulting signal is still considered "raw ' . because all information is treated equally
- the pertormance ot the artificial neural network is dependent on the data that is used to train the network
- the amount ot data and the distribution ot the data w ithin the realm ot possibilities are known to hav e a large effect on the ability ot the network to recognize patterns and to generalize
- Data tor the OSS is made up of v ectors
- Each v ector is a combination of the useful parts of the signals collected from four ultrasonic transducers
- a ty pical v ector could comprise on the order of 100 data points, each representing the (time displaced) echo lev el as recorded by the ultrasonic transducers
- the first set contains the patterns that the ANN is being trained on to recognize as either an airbag deploy or non-deploy scenario
- the second set is the independent test data This set is used during the network training to direct the optimization of the network weights
- the third set is the v alidation (or real world) data This set is used to quantify the success rate (or performance) of the finalized artificial neural network
- Table 1 shows the main characteristics ot these three data sets as collected lor the v ehicle
- Three numbers characterize the sets T he number ot configurations characterizes how manv di f ferent subjects and objectss were used
- the number ot setups is the product ot the number ot configurations and the number ot v ehicle interior v ariations ( seat position and recline roof and w indow state, etc ) performed for each configuration
- the total number ot v ectors is then made up ot the product of the number ot setups and the number ot patterns collected while the subject or obiect mov es w ithin the passenger v olume
- the training data set can be split up in v arious wav s into subsets that show the distribution of the data
- Table 2 shows the distribution ot the training set amongst three classes ot passenger seat occupancy Empty Seat Human Occupant and Child Seat All human occupants were adults of v arious sizes No children were part of the training data set other then those seated in 1 orward Facing Child Seats
- Table 3 shows a further breakup ot the Child Seats into Forward racing Child Seats Rearward Facing Child Seats.
- Table 4 shows a different tv pe ot distribution one based on the env ironmental conditions inside the v ehicle
- the independent test data is created using the same configurations subjects objects and conditions as used for the training data set Its makeup and distributions are theretore the same as those ot the training data set
- the baseline network consisted ot a tour laver back-propagation network with 117 input layer nodes 20 and 7 nodes respectively in the two hidden layers and 1 output laver node
- the input laver is made up ot inputs from four ultrasonic transducers These were located in the vehicle on the rear quarter panel (A) the A-pillar (B) and the over-head console (C II) lable 9 shows the number of points taken trom each of these channels that make up one v ectoi Table 9 Transducer Volume
- the artificial neural network is implemented using the NeuralW orks Protessional II/PIus software
- the method used for training the decision mathematical model was back-propagation with Extended Delta- Bar-Delta learning rule and sigmoid transfer function
- the Extended DBD paradigm uses past values of the gradient to infer the local curv ature of the error surface This leads to a learning rule in which every connection has a different learning rate and a ditferent momentum term both ot which are automatically calculated
- the network was trained using the above-described training and independent test data sets An optimum (against the independent test set) was found after 3 675 000 training cv cles Each training cycle uses 30 v ectors (known as the epoch) randomly chosen from the 650 000 av ailable training set vectors Table 10 shows the performance of the baseline network
- Normalization Normalization is used to scale the leal world data range into a range acceptable tor the network training
- the NeuralWorks software requires the use ot a scaling factor to bring the input data into a range ot 0 to 1 inclusive
- Several normalization methods have been explored tor their effect on the svstem performance
- the real world data consists of 12 bit digitized signals with values between 0 and 4095
- FIG 20 shows a tvpical raw signal
- a raw vector consists ot combined sections ot four signals
- ⁇ n 11 is the change in the network weights lcoef the learning coefficient e 1 ' is the local error at neuron I in laver s ' is the current output state ot neuron ⁇ in laver s
- the baseline network is a back-propagation type network
- Back-propagation is a general -purpose network paradigm that has been successfully used for prediction, classification system modeling, and filtering as well as manv other general tv pes ot problems
- Back propagation learns by calculating an error between desired and actual output and propagating this error information back to each node in the network
- This back-propagated error is used to driv e the learning at each node
- Some ot the adv antages ot a back- propagation network are that it attempts to minimize the global error and that it can prov ide a v erv compact distributed representation of complex data sets
- Some of the disadv antages are its slow learning and the irregular boundaries and unexpected classification regions due to the distributed nature ot the network and the use of a transfer functions that is unbounded
- Some of these disadv antages can be ov ercome bv using a modified back-propagation method sucn as the
- T he spatial distribution of the independent test data was as w ide as that of the training data This has resulted in a network that can generalize across a large spatial v olume A higher performance across a smaller volume, located immediately around the peak of the normal distribution, combined w ith a lower performance on the outskirts ot the distribution curve, might be preferable
- the distribution ot the independent test set needs to be a reflection of the normal distribution for the system (a k.a. native population)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2001519213A JP2004522932A (en) | 1999-05-27 | 2000-05-30 | Method for developing a system for identifying the presence and orientation of an object in a vehicle |
DE10084638T DE10084638T1 (en) | 1999-05-27 | 2000-05-30 | Method for developing the system by identifying the presence and location of the object in the vehicle |
SE0100186A SE523753C2 (en) | 1999-05-27 | 2001-01-24 | Method for developing a system for identifying the presence and position of an object in a vehicle |
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13616399P | 1999-05-27 | 1999-05-27 | |
US60/136,163 | 1999-05-27 | ||
US09/382,406 | 1999-08-24 | ||
US09/382,406 US6529809B1 (en) | 1997-02-06 | 1999-08-24 | Method of developing a system for identifying the presence and orientation of an object in a vehicle |
US09/474,147 US6397136B1 (en) | 1997-02-06 | 1999-12-29 | System for determining the occupancy state of a seat in a vehicle |
US09/474,147 | 1999-12-29 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2001014910A2 true WO2001014910A2 (en) | 2001-03-01 |
WO2001014910A3 WO2001014910A3 (en) | 2001-09-27 |
Family
ID=27384821
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2000/014903 WO2001014910A2 (en) | 1999-05-27 | 2000-05-30 | Method for developing a system for identifying the presence and orientation of an object in a vehicle |
Country Status (5)
Country | Link |
---|---|
US (2) | US6397136B1 (en) |
JP (1) | JP2004522932A (en) |
DE (1) | DE10084638T1 (en) |
SE (1) | SE523753C2 (en) |
WO (1) | WO2001014910A2 (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003006279A1 (en) * | 2001-07-10 | 2003-01-23 | Siemens Aktiengesellschaft | System for monitoring the interior of a vehicle |
ES2222784A1 (en) * | 2002-10-17 | 2005-02-01 | Proyectos Y Tecnologia Sallen, S.L. | Detector of e.g. explosives under a motor vehicle includes a strain gauge with an electronic controller and anchors fixed to the vehicle |
DE102004018288A1 (en) * | 2004-04-15 | 2005-11-03 | Conti Temic Microelectronic Gmbh | Approximate identification of object involves determining characteristics in defined region from graphic data in lines, columns or matrix form, determining numeric approximate value for object from graphic data by statistical classification |
DE10239761B4 (en) * | 2002-08-29 | 2007-10-25 | Sartorius Ag | Method and device for identifying the type of occupancy of a support surface |
US7436299B2 (en) | 2001-03-02 | 2008-10-14 | Elesys North America Inc. | Vehicle occupant detection using relative impedance measurements |
JP2008302930A (en) * | 2001-03-02 | 2008-12-18 | Elesys North America Inc | Multiple sensor vehicle occupant detection system and method for airbag deployment control |
CN104369739A (en) * | 2013-08-12 | 2015-02-25 | 通用汽车环球科技运作有限责任公司 | Vehicle systems and methods for identifying a driver |
WO2016164793A1 (en) * | 2015-04-10 | 2016-10-13 | Robert Bosch Gmbh | Detection of occupant size and pose with a vehicle interior camera |
CN106364431A (en) * | 2015-07-24 | 2017-02-01 | 福特环球技术公司 | Device and method for adjusting comfort of chair of at least one people inside vehicle |
CN107438808A (en) * | 2016-10-31 | 2017-12-05 | 深圳市大疆创新科技有限公司 | A kind of method, apparatus and relevant device of rod volume control |
CN110576818A (en) * | 2018-06-11 | 2019-12-17 | 沃尔沃汽车公司 | method and system for controlling the state of an occupant protection feature of a vehicle |
CN111942284A (en) * | 2020-07-06 | 2020-11-17 | 山东师范大学 | System and method for detecting and alarming people left in vehicle |
Families Citing this family (90)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080234899A1 (en) * | 1992-05-05 | 2008-09-25 | Automotive Technologies International, Inc. | Vehicular Occupant Sensing and Component Control Techniques |
US7660437B2 (en) * | 1992-05-05 | 2010-02-09 | Automotive Technologies International, Inc. | Neural network systems for vehicles |
US7766383B2 (en) * | 1998-11-17 | 2010-08-03 | Automotive Technologies International, Inc. | Vehicular component adjustment system and method |
US20070135982A1 (en) * | 1995-06-07 | 2007-06-14 | Automotive Technologies International, Inc. | Methods for Sensing Weight of an Occupying Item in a Vehicular Seat |
JP3532772B2 (en) * | 1998-09-25 | 2004-05-31 | 本田技研工業株式会社 | Occupant state detection device |
IT1305155B1 (en) * | 1998-11-03 | 2001-04-10 | Valeo Sicurezza Abitacolo Spa | HANDLE FOR A VEHICLE DOOR. |
US6578870B2 (en) * | 2000-07-12 | 2003-06-17 | Siemens Ag | Vehicle occupant weight classification system |
US6539284B2 (en) * | 2000-07-25 | 2003-03-25 | Axonn Robotics, Llc | Socially interactive autonomous robot |
US6801662B1 (en) * | 2000-10-10 | 2004-10-05 | Hrl Laboratories, Llc | Sensor fusion architecture for vision-based occupant detection |
US6836754B2 (en) * | 2001-01-31 | 2004-12-28 | Key Safety Systems, Inc. | Biomechanical system development of a restraint system |
EP1390844A4 (en) * | 2001-04-24 | 2006-01-18 | Breed Automotive Tech | Simplified modeling software interface and method |
US6925193B2 (en) * | 2001-07-10 | 2005-08-02 | Eaton Corporation | Image processing system for dynamic suppression of airbags using multiple model likelihoods to infer three dimensional information |
US6542802B2 (en) * | 2001-07-02 | 2003-04-01 | Delphi Technologies, Inc. | Vehicle occupant characterization method with rough road compensation |
DE10163910A1 (en) * | 2001-12-22 | 2003-07-03 | Bosch Gmbh Robert | Occupant classification device and method for occupant classification |
US7026946B2 (en) * | 2002-04-17 | 2006-04-11 | Darrel Saunders | Method and apparatus for sensing seat occupancy |
US7046158B2 (en) * | 2002-04-17 | 2006-05-16 | Darrel Saunders | Method and apparatus for sensing seat occupancy |
GB0211566D0 (en) * | 2002-05-21 | 2002-06-26 | Koninkl Philips Electronics Nv | Method and apparatus for providing travel relating information to a user |
US6650978B1 (en) * | 2002-05-31 | 2003-11-18 | Delphi Technologies, Inc. | Method and apparatus for preserving calibration data in a vehicle seat occupant detection system |
JP3922637B2 (en) * | 2002-08-30 | 2007-05-30 | 本田技研工業株式会社 | Side airbag system |
US7177743B2 (en) * | 2003-06-02 | 2007-02-13 | Toyota Engineering & Manufacturing North America, Inc. | Vehicle control system having an adaptive controller |
US7075450B2 (en) * | 2003-06-26 | 2006-07-11 | Lear Corporation | Vehicle occupant sensing system having discrete wiring |
US7172244B2 (en) * | 2003-06-26 | 2007-02-06 | Lear Corporation | Vehicle seat assembly having a vehicle occupant sensing system and a seat cushion insert positioned therein |
US6994397B2 (en) * | 2003-06-26 | 2006-02-07 | Lear Corporation | Vehicle occupant sensing system having sensor assemblies with variable blasing member |
US7063382B2 (en) * | 2003-06-26 | 2006-06-20 | Lear Corporation | Vehicle seat assembly having a vehicle occupant sensing system and a seat cushion insert |
US7258398B2 (en) * | 2003-06-26 | 2007-08-21 | Lear Corporation | Vehicle occupant sensing system having an upper slide member with an emitter interference member |
US7132953B2 (en) | 2003-06-26 | 2006-11-07 | Lear Corporation | Spring sensor assembly for a vehicle seat cushion |
US7043997B2 (en) * | 2003-07-09 | 2006-05-16 | Cherry Corporation | Seat for sensing a load |
US7239724B2 (en) * | 2003-07-22 | 2007-07-03 | International Business Machines Corporation | Security identification system and method |
US7035432B2 (en) * | 2003-07-22 | 2006-04-25 | Ronjo Company | Method of monitoring sleeping infant |
KR100513879B1 (en) * | 2003-08-08 | 2005-09-09 | 현대자동차주식회사 | Method for Expansion Pressure Control of Assistant Seat Airbag |
US7151452B2 (en) * | 2003-12-05 | 2006-12-19 | Elesys North America Inc. | Vehicle occupant sensing system |
US7059029B2 (en) | 2003-12-30 | 2006-06-13 | Lear Corporation | Method of testing a sensor array incorporated into a vehicle seat |
US6901322B1 (en) | 2003-12-30 | 2005-05-31 | Lear Corporation | Method of predicting an empty seat condition in an occupancy sensing system |
US7034670B2 (en) * | 2003-12-30 | 2006-04-25 | Lear Corporation | Method of occupancy classification in a vehicle seat |
US7053759B2 (en) * | 2003-12-30 | 2006-05-30 | Lear Corporation | Method of determining an equivalent value for a failed sensor in a vehicle seat having an occupancy sensing system |
US6985077B2 (en) * | 2003-12-30 | 2006-01-10 | Lear Corporation | Method of tuning a sensor array for occupancy sensing in a vehicle seat |
US7185916B2 (en) * | 2004-01-14 | 2007-03-06 | Lear Corporation | Vehicle seat assembly having a field effect sensor for detecting seat position |
DE102004016185A1 (en) * | 2004-04-01 | 2005-10-20 | Volkswagen Ag | Manufacturing component, for motor vehicle, involves determining movement of test body relative to independent coordinate system depending on measurement values for translational movement of movement measurement device |
WO2005120890A2 (en) * | 2004-06-07 | 2005-12-22 | Delphi Technologies, Inc. | Child seat monitoring system and method for determining a type of child seat |
US7225067B2 (en) * | 2004-07-02 | 2007-05-29 | Lear Corporation | Vehicle occupant sensing system for a vehicle seat assembly and method of operating the same |
JP2006023868A (en) * | 2004-07-07 | 2006-01-26 | Toshiba Corp | Solution searching device and initial value setting method |
US8041482B2 (en) * | 2004-07-29 | 2011-10-18 | Iee International Electronics & Engineering S.A. | Seat sensor system |
WO2006020604A2 (en) * | 2004-08-09 | 2006-02-23 | Delphi Technologies, Inc. | Child restraint system comprising event data recorder |
US7288009B2 (en) * | 2004-09-08 | 2007-10-30 | Delphi Technologies, Inc. | Apparatus and method for interconnecting a child seat and monitoring system |
US7428942B2 (en) * | 2004-10-27 | 2008-09-30 | Lear Corporation | Vehicle occupant sensing system having guiding ribs |
US7365278B2 (en) * | 2004-10-27 | 2008-04-29 | Lear Corporation | Vehicle occupant sensing system having a contamination barrier member |
US7402769B2 (en) * | 2004-10-27 | 2008-07-22 | Lear Corporation | Vehicle occupant sensing system having a retention member for a biasing member |
US20060097497A1 (en) * | 2004-10-27 | 2006-05-11 | Sallam Faisal K | Vehicle occupant sensing system having a contamination barrier member |
US7405370B2 (en) * | 2004-10-27 | 2008-07-29 | Lear Corporation | Vehicle occupant sensing system having enclosed sensor assembly |
US7100980B2 (en) * | 2004-10-27 | 2006-09-05 | Lear Corporation | Vehicle seat assembly having a vehicle occupant sensing system with a biasing pad |
US20060273897A1 (en) * | 2005-06-03 | 2006-12-07 | Risi Alan | Dynamic software system for a security checkpoint |
US7505841B2 (en) * | 2005-09-02 | 2009-03-17 | Delphi Technologies, Inc. | Vision-based occupant classification method and system for controlling airbag deployment in a vehicle restraint system |
US20070285218A1 (en) * | 2006-06-09 | 2007-12-13 | 3M Innovative Properties Company | Occupant abandonment sensor for automotive vehicles |
US7656169B2 (en) * | 2007-02-06 | 2010-02-02 | Iee International Electronics & Engineering S.A. | Capacitive occupant detection system |
WO2008109801A1 (en) * | 2007-03-07 | 2008-09-12 | Kmt Robotic Solutions, Inc. | System and method of locating relative positions of objects |
US7995816B2 (en) * | 2007-09-24 | 2011-08-09 | Baxter International Inc. | Detecting access disconnect by pattern recognition |
US8384399B2 (en) | 2008-08-28 | 2013-02-26 | Infineon Technologies Ag | System including capacitively coupled electrodes and circuits in a network |
US20100191390A1 (en) * | 2009-01-28 | 2010-07-29 | Delphi Technologies, Inc. | System and method for detecting the occupancy of a vehicle seat |
FR2943236A1 (en) * | 2009-03-18 | 2010-09-24 | Imra Europ Sas | METHOD FOR MONITORING A BIOLOGICAL PARAMETER OF A PERSON USING SENSORS |
US20100304762A1 (en) * | 2009-05-27 | 2010-12-02 | Bernard Joseph Hall | Indoor tracking system |
US20110190987A1 (en) * | 2010-02-04 | 2011-08-04 | Delphi Technologies, Inc. | Occupant detection system and method |
US20110190980A1 (en) * | 2010-02-04 | 2011-08-04 | Delphi Technologies, Inc. | Occupant detection system and method |
US11857331B1 (en) | 2013-01-19 | 2024-01-02 | Bertec Corporation | Force measurement system |
US11311209B1 (en) | 2013-01-19 | 2022-04-26 | Bertec Corporation | Force measurement system and a motion base used therein |
US11540744B1 (en) | 2013-01-19 | 2023-01-03 | Bertec Corporation | Force measurement system |
US11052288B1 (en) * | 2013-01-19 | 2021-07-06 | Bertec Corporation | Force measurement system |
US9518860B2 (en) * | 2013-01-31 | 2016-12-13 | Apm Automation Solutions Ltd | System and method for fuzzy logic based measurement of a content of a bin |
US9440608B2 (en) * | 2013-03-13 | 2016-09-13 | Aisin Seiki Kabushiki Kaisha | Seat occupant determining apparatus |
US9651649B1 (en) * | 2013-03-14 | 2017-05-16 | The Trustees Of The Stevens Institute Of Technology | Passive acoustic detection, tracking and classification system and method |
WO2014146623A1 (en) | 2013-03-19 | 2014-09-25 | BALLUF GmbH | Inductive displacement measuring sensor and method for operating the latter |
EP2808693B1 (en) * | 2013-05-27 | 2019-08-14 | Volvo Car Corporation | System and method for determining a position of a living being in a vehicle |
US10542961B2 (en) | 2015-06-15 | 2020-01-28 | The Research Foundation For The State University Of New York | System and method for infrasonic cardiac monitoring |
CN106897310B (en) * | 2015-12-18 | 2020-04-17 | 英业达科技有限公司 | Vehicle personalized configuration adjustment system and method thereof |
EP3479352A4 (en) * | 2016-06-29 | 2020-01-08 | Seeing Machines Limited | Camera registration in a multi-camera system |
US10676058B2 (en) | 2016-07-06 | 2020-06-09 | Ford Global Technologies, Llc | Vehicle dashboard safety features |
CN107766805A (en) | 2017-09-29 | 2018-03-06 | 阿里巴巴集团控股有限公司 | Lift the method, apparatus and server of car damage identification image recognition result |
US11180154B2 (en) | 2017-10-17 | 2021-11-23 | The Regents Of The University Of Michigan | Fingerprinting drivers based on vehicle turns |
CN107562213A (en) * | 2017-10-27 | 2018-01-09 | 网易(杭州)网络有限公司 | Detection method, device and the wear-type visual device of visual fatigue state |
KR102016149B1 (en) * | 2017-11-15 | 2019-08-29 | 주식회사 서연전자 | Driver sensing device and method thereof |
KR102567188B1 (en) | 2017-12-20 | 2023-08-16 | 현대자동차주식회사 | Vehicle and controlling method of the same |
US11925446B2 (en) * | 2018-02-22 | 2024-03-12 | Vayyar Imaging Ltd. | Radar-based classification of vehicle occupants |
CN108482302B (en) * | 2018-03-09 | 2020-08-04 | 北京汽车股份有限公司 | Safety belt state reminding system, reminding control method and vehicle |
US10726310B2 (en) * | 2018-03-31 | 2020-07-28 | Veoneer Us Inc. | Deployment zone definition and associated restraint control |
US11230293B2 (en) * | 2018-06-11 | 2022-01-25 | Iee International Electronics & Engineering S.A. | Method for robust vehicle occupancy detection with vital sign monitoring |
CN111422200B (en) * | 2019-01-09 | 2022-02-08 | 北京地平线机器人技术研发有限公司 | Method and device for adjusting vehicle equipment and electronic equipment |
DE102019204157A1 (en) * | 2019-03-26 | 2020-10-01 | Brose Fahrzeugteile Se & Co. Kommanditgesellschaft, Bamberg | Method for monitoring the interior of a vehicle, monitoring arrangement and vehicle |
JPWO2021256303A1 (en) * | 2020-06-18 | 2021-12-23 | ||
CN112193251B (en) * | 2020-10-13 | 2022-04-05 | 一汽解放汽车有限公司 | Driving posture feedback system and method for driver |
DE112021007659T5 (en) | 2021-05-13 | 2024-03-14 | Mitsubishi Electric Corporation | Vehicle occupant recognition system and vehicle occupant recognition method |
DE102022126044A1 (en) | 2022-10-10 | 2024-04-25 | HELLA GmbH & Co. KGaA | Method for operating a control device, control device and vehicle |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5482314A (en) * | 1994-04-12 | 1996-01-09 | Aerojet General Corporation | Automotive occupant sensor system and method of operation by sensor fusion |
Family Cites Families (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3275975A (en) | 1964-12-21 | 1966-09-27 | Cleveland Technical Ct Inc | Ultrasonic detecting means |
US5829782A (en) | 1993-03-31 | 1998-11-03 | Automotive Technologies International, Inc. | Vehicle interior identification and monitoring system |
US5845000A (en) | 1992-05-05 | 1998-12-01 | Automotive Technologies International, Inc. | Optical identification and monitoring system using pattern recognition for use with vehicles |
US4810100A (en) * | 1982-07-19 | 1989-03-07 | Honeywell Inc. | Ultrasonic energy transfer sensing system |
JPS6060838A (en) | 1983-09-13 | 1985-04-08 | 富士通株式会社 | Ultrasonic sound field observing apparatus |
US4881270A (en) | 1983-10-28 | 1989-11-14 | The United States Of America As Represented By The Secretary Of The Navy | Automatic classification of images |
US4906940A (en) | 1987-08-24 | 1990-03-06 | Science Applications International Corporation | Process and apparatus for the automatic detection and extraction of features in images and displays |
JPH01158579A (en) | 1987-09-09 | 1989-06-21 | Aisin Seiki Co Ltd | Image recognizing device |
EP0357225B1 (en) | 1988-07-29 | 1993-12-15 | Mazda Motor Corporation | Air bag system for automobile |
JP2878409B2 (en) | 1989-09-04 | 1999-04-05 | 株式会社リコー | 3D object imaging method |
US5071160A (en) | 1989-10-02 | 1991-12-10 | Automotive Systems Laboratory, Inc. | Passenger out-of-position sensor |
DE4005598C2 (en) | 1990-02-22 | 2000-06-15 | Bosch Gmbh Robert | Protection procedure for vehicle occupants and device for carrying out the procedure |
JP2605922B2 (en) | 1990-04-18 | 1997-04-30 | 日産自動車株式会社 | Vehicle safety devices |
US5214744A (en) | 1990-12-14 | 1993-05-25 | Westinghouse Electric Corp. | Method and apparatus for automatically identifying targets in sonar images |
US5181254A (en) | 1990-12-14 | 1993-01-19 | Westinghouse Electric Corp. | Method for automatically identifying targets in sonar images |
US5684701A (en) | 1995-06-07 | 1997-11-04 | Automotive Technologies International, Inc. | Method and apparatus for sensing a vehicle crash |
US5305012A (en) | 1992-04-15 | 1994-04-19 | Reveo, Inc. | Intelligent electro-optical system and method for automatic glare reduction |
US6081757A (en) * | 1995-06-07 | 2000-06-27 | Automotive Technologies International, Inc. | Seated-state detecting apparatus |
US6078854A (en) * | 1995-06-07 | 2000-06-20 | Automotive Technologies International, Inc. | Apparatus and method for adjusting a vehicle component |
US5943295A (en) | 1997-02-06 | 1999-08-24 | Automotive Technologies International Inc. | Method for identifying the presence and orientation of an object in a vehicle |
US5235339A (en) | 1992-11-13 | 1993-08-10 | The United States Of America As Represented By The Secretary Of The Navy | Radar target discrimination systems using artificial neural network topology |
US5330226A (en) | 1992-12-04 | 1994-07-19 | Trw Vehicle Safety Systems Inc. | Method and apparatus for detecting an out of position occupant |
US5298732A (en) | 1993-02-18 | 1994-03-29 | Emee, Inc. | Automatic visor for continuously repositioning a shading object to shade a designated location from a direct radiation source |
WO1994022693A1 (en) | 1993-03-31 | 1994-10-13 | Automotive Technologies International, Inc. | Vehicle occupant position and velocity sensor |
US5446661A (en) | 1993-04-15 | 1995-08-29 | Automotive Systems Laboratory, Inc. | Adjustable crash discrimination system with occupant position detection |
US5390136A (en) | 1993-06-14 | 1995-02-14 | Motorola, Inc. | Artificial neuron and method of using same |
US5366241A (en) | 1993-09-30 | 1994-11-22 | Kithil Philip W | Automobile air bag system |
US5605348A (en) | 1993-11-03 | 1997-02-25 | Trw Vehicle Safety Systems Inc. | Method and apparatus for sensing a rearward facing child seat |
US5454591A (en) | 1993-11-03 | 1995-10-03 | Trw Vehicle Safety Systems Inc. | Method and apparatus for sensing a rearward facing child restraining seat |
US5413378A (en) | 1993-12-02 | 1995-05-09 | Trw Vehicle Safety Systems Inc. | Method and apparatus for controlling an actuatable restraining device in response to discrete control zones |
US5626359A (en) | 1993-12-02 | 1997-05-06 | Trw Vehicle Safety Systems, Inc. | Method and apparatus for controlling an actuatable restraining device in response to discrete control zones |
JPH07196006A (en) | 1993-12-29 | 1995-08-01 | Sensor Technol Kk | Occupant detection system for front passenger seat |
DE4400664C2 (en) | 1994-01-12 | 1995-11-16 | Spies Martin Dipl Ing Fh | Device for detecting seat occupancy in vehicles etc. |
US5602734A (en) | 1994-09-23 | 1997-02-11 | Advanced Safety Concepts, Inc. | Automobile air bag systems |
JP3494769B2 (en) | 1994-12-20 | 2004-02-09 | 株式会社東海理化電機製作所 | Airbag device for passenger or rear seat |
JP2973863B2 (en) | 1995-03-31 | 1999-11-08 | トヨタ自動車株式会社 | Airbag device for passenger seat |
US5722686A (en) | 1995-05-16 | 1998-03-03 | Trw Vehicle Safety Systems, Inc. | Method and apparatus for sensing an occupant position using capacitance sensing |
US5782482A (en) | 1996-01-30 | 1998-07-21 | K-2 Corporation | Snowboard and method of construction |
US6007095A (en) | 1997-02-05 | 1999-12-28 | Automotive Systems Laboratory, Inc. | Vehicle occupant position sensor |
US5954360A (en) | 1997-09-18 | 1999-09-21 | Breed Automotive Technology, Inc. | Vehicle occupant sensing apparatus and method |
-
1999
- 1999-12-29 US US09/474,147 patent/US6397136B1/en not_active Expired - Fee Related
-
2000
- 2000-05-30 DE DE10084638T patent/DE10084638T1/en not_active Withdrawn
- 2000-05-30 JP JP2001519213A patent/JP2004522932A/en not_active Withdrawn
- 2000-05-30 WO PCT/US2000/014903 patent/WO2001014910A2/en active Application Filing
-
2001
- 2001-01-24 SE SE0100186A patent/SE523753C2/en unknown
- 2001-02-09 US US09/780,654 patent/US6459973B1/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5482314A (en) * | 1994-04-12 | 1996-01-09 | Aerojet General Corporation | Automotive occupant sensor system and method of operation by sensor fusion |
US5890085A (en) * | 1994-04-12 | 1999-03-30 | Robert Bosch Corporation | Methods of occupancy state determination and computer programs |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7436299B2 (en) | 2001-03-02 | 2008-10-14 | Elesys North America Inc. | Vehicle occupant detection using relative impedance measurements |
JP2008302930A (en) * | 2001-03-02 | 2008-12-18 | Elesys North America Inc | Multiple sensor vehicle occupant detection system and method for airbag deployment control |
WO2003006279A1 (en) * | 2001-07-10 | 2003-01-23 | Siemens Aktiengesellschaft | System for monitoring the interior of a vehicle |
US7124007B2 (en) | 2001-07-10 | 2006-10-17 | Siemens Aktiengesellschaft | System for monitoring the interior of a vehicle |
KR100862109B1 (en) * | 2001-07-10 | 2008-10-09 | 지멘스 악티엔게젤샤프트 | System for monitoring the interior of a vehicle |
DE10239761B4 (en) * | 2002-08-29 | 2007-10-25 | Sartorius Ag | Method and device for identifying the type of occupancy of a support surface |
ES2222784A1 (en) * | 2002-10-17 | 2005-02-01 | Proyectos Y Tecnologia Sallen, S.L. | Detector of e.g. explosives under a motor vehicle includes a strain gauge with an electronic controller and anchors fixed to the vehicle |
DE102004018288A1 (en) * | 2004-04-15 | 2005-11-03 | Conti Temic Microelectronic Gmbh | Approximate identification of object involves determining characteristics in defined region from graphic data in lines, columns or matrix form, determining numeric approximate value for object from graphic data by statistical classification |
CN104369739A (en) * | 2013-08-12 | 2015-02-25 | 通用汽车环球科技运作有限责任公司 | Vehicle systems and methods for identifying a driver |
WO2016164793A1 (en) * | 2015-04-10 | 2016-10-13 | Robert Bosch Gmbh | Detection of occupant size and pose with a vehicle interior camera |
US10829072B2 (en) | 2015-04-10 | 2020-11-10 | Robert Bosch Gmbh | Detection of occupant size and pose with a vehicle interior camera |
CN106364431A (en) * | 2015-07-24 | 2017-02-01 | 福特环球技术公司 | Device and method for adjusting comfort of chair of at least one people inside vehicle |
CN107438808A (en) * | 2016-10-31 | 2017-12-05 | 深圳市大疆创新科技有限公司 | A kind of method, apparatus and relevant device of rod volume control |
CN110576818A (en) * | 2018-06-11 | 2019-12-17 | 沃尔沃汽车公司 | method and system for controlling the state of an occupant protection feature of a vehicle |
EP3581440A1 (en) * | 2018-06-11 | 2019-12-18 | Volvo Car Corporation | Method and system for controlling a state of an occupant protection feature for a vehicle |
CN111942284A (en) * | 2020-07-06 | 2020-11-17 | 山东师范大学 | System and method for detecting and alarming people left in vehicle |
Also Published As
Publication number | Publication date |
---|---|
US6397136B1 (en) | 2002-05-28 |
JP2004522932A (en) | 2004-07-29 |
US20020082756A1 (en) | 2002-06-27 |
SE523753C2 (en) | 2004-05-11 |
WO2001014910A3 (en) | 2001-09-27 |
US6459973B1 (en) | 2002-10-01 |
SE0100186D0 (en) | 2001-01-24 |
SE0100186L (en) | 2001-03-27 |
DE10084638T1 (en) | 2002-05-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6397136B1 (en) | System for determining the occupancy state of a seat in a vehicle | |
US6452870B1 (en) | Methods for controlling deployment of an occupant restraint in a vehicle and determining whether the occupant is a child seat | |
US6445988B1 (en) | System for determining the occupancy state of a seat in a vehicle and controlling a component based thereon | |
US6529809B1 (en) | Method of developing a system for identifying the presence and orientation of an object in a vehicle | |
US6757602B2 (en) | System for determining the occupancy state of a seat in a vehicle and controlling a component based thereon | |
US5943295A (en) | Method for identifying the presence and orientation of an object in a vehicle | |
USRE37260E1 (en) | Method for identifying the presence and orientation of an object in a vehicle | |
US7712777B2 (en) | Airbag deployment control based on contact with occupant | |
US7620521B2 (en) | Dynamic weight sensing and classification of vehicular occupants | |
US7762582B2 (en) | Vehicle component control based on occupant morphology | |
US7900736B2 (en) | Vehicular seats with fluid-containing weight sensing system | |
US7779956B2 (en) | Vehicular seats with weight sensing capability | |
US7359527B2 (en) | Combined occupant weight and spatial sensing in a vehicle | |
US7147246B2 (en) | Method for airbag inflation control | |
US7770920B2 (en) | Vehicular seats with fluid-containing weight sensing system | |
US8820782B2 (en) | Arrangement for sensing weight of an occupying item in vehicular seat | |
US7650212B2 (en) | Pedal adjustment system and method | |
US7976060B2 (en) | Seat load or displacement measuring system for occupant restraint system control | |
US20070132220A1 (en) | Occupant Classification and Airbag Deployment Suppression Based on Weight | |
US7988190B2 (en) | Airbag deployment control using seatbelt-mounted sensor | |
US7575248B2 (en) | Airbag inflation control based on sensed contact with occupant | |
US20080142713A1 (en) | Vehicular Occupant Sensing Using Infrared | |
US20080189053A1 (en) | Apparatus and Method for Analyzing Weight of an Occupying Item of a Vehicular Seat | |
Breed et al. | Development of an occupant position sensor system to improve frontal crash protection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 01001866 Country of ref document: SE |
|
AK | Designated states |
Kind code of ref document: A2 Designated state(s): DE GB JP SE |
|
AL | Designated countries for regional patents |
Kind code of ref document: A2 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
WWP | Wipo information: published in national office |
Ref document number: 01001866 Country of ref document: SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
AK | Designated states |
Kind code of ref document: A3 Designated state(s): DE GB JP SE |
|
AL | Designated countries for regional patents |
Kind code of ref document: A3 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
REG | Reference to national code |
Ref country code: DE Ref legal event code: 8642 |
|
ENP | Entry into the national phase |
Ref document number: 2001 519213 Country of ref document: JP Kind code of ref document: A |
|
RET | De translation (de og part 6b) |
Ref document number: 10084638 Country of ref document: DE Date of ref document: 20020502 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10084638 Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |