Procedure for detecting seat occupancy

Abstract

Methods for detecting seat occupancy in a motor vehicle comprising at least the following steps: a) Establishing a testing sequence for the seats, at least using a probability of the seats being occupied, b) Sequential check of the seat occupancy using the check sequence created in step a).

Description

State of the art

[0001] The present invention relates to a method for detecting seat occupancy, particularly of seats in a motor vehicle. Seat occupancy detection can be taken into account when triggering safety systems, particularly in the event of a collision between the motor vehicle and a collision object. Thus, seat occupancy information can contribute to the safety of the vehicle's occupants.

[0002] Modern vehicles regularly feature a warning function that detects whether a passenger is not wearing a seatbelt. This not only checks whether a seatbelt buckle is closed, but also evaluates other information to determine whether a seat is occupied. Only if this is the case will a warning signal be issued if the seatbelt buckle is not closed. Furthermore, seat occupancy information is considered when deploying safety systems such as airbags.

[0003] However, known methods for detecting seat occupancy in a motor vehicle usually require a very large amount of computing effort and / or expensive sensor technology. Disclosure of the invention

[0004] According to claim 1, a particularly advantageous method for detecting seat occupancy in a motor vehicle is disclosed. The dependent claims specify particularly advantageous developments of the method.

[0005] Seat occupancy here refers specifically to information about whether a person is sitting on a seat or not. Furthermore, the term "seat occupancy" as defined here also includes information about the type of seat occupancy, such as the weight of a person sitting on a seat. This allows, in particular, a distinction to be made between adults, children, infant carriers, and objects on a seat.

[0006] The safety of the vehicle's occupants can be increased by knowing which seat in a motor vehicle is occupied (and how that seat is occupied). For example, a warning signal can be issued if a seat is detected as occupied but the seat belt buckle for that seat is not fastened. Seat occupancy can also be taken into account when deploying restraint systems in the event of an accident. In particular, information about the type of seat occupancy can be used. If, for example, a child is sitting on a seat, restraint systems can be deployed differently than if an adult is sitting in the seat. If, for example, a baby seat is detected on the front passenger seat, the deployment of a front passenger airbag is preferentially suppressed. The method described can make it possible to detect seat occupancy particularly quickly using particularly low computing effort.The described method can in particular contribute to the particularly efficient use of computing capacity.

[0007] According to the described method, the occupancy of the individual seats of the motor vehicle is preferably checked sequentially. It is particularly preferred that a seat is only checked after a previous check of another seat has been completed (unless it is the first seat checked). During such a check, a check result for a seat can be output and / or further processed before the entire check is completed. This allows partial check results to be obtained particularly quickly. The seats are checked in the check sequence established in step a).

[0008] In step a), occupancy probabilities of the seats to be tested are preferably provided first. The motor vehicle seats are preferably generally ordered in descending order of occupancy probability, which establishes the test sequence. However, deviations from this specified test sequence are possible. In particular, it is preferable not to classify seats with a very high occupancy probability (especially seats with an occupancy probability above a specified threshold) according to occupancy probability, but rather to place them at the end of the test sequence, contrary to the arrangement of the seats according to descending occupancy probability. This applies in particular to the driver's seat, which is always occupied in conventional motor vehicles. The driver's seat is preferably tested last or not tested at all.In many motor vehicles, for example, the driver's seat, unlike the other seats, does not have a weight sensor that could be used to detect seat occupancy because the driver's seat can be assumed to be occupied throughout the entire operation of the motor vehicle. A corresponding warning signal for an unfastened seat belt is preferably issued if the driver's seat belt buckle is not closed. In this case, a check of the occupancy of the driver's seat is preferably not waited for. In motor vehicles that enable at least partially autonomous driving (i.e. without intervention by a driver), the driver's seat does not have to be occupied during the entire operation of the motor vehicle. In this case, it is preferable that the driver's seat is taken into account in the test sequence like the other seats and, in particular, that the driver's seat is tested first.

[0009] When establishing the test sequence, the importance of knowing the occupancy of each seat can also be taken into account. For example, there may be a legal requirement that the occupancy of the front passenger seat must be detected no later than a specified time after the vehicle was put into operation. Such a regulation can be used in particular to suppress the deployment of a passenger airbag if there is a baby seat on the front passenger seat. Knowing the occupancy of the front passenger seat can therefore be particularly important. In the case of the driver's seat, on the other hand, it can be assumed that it is always occupied. Knowing the occupancy of the driver's seat can therefore be considered less important because it is already known as a virtually certain assumption.

[0010] Furthermore, it is preferable that for each seat (or at least for a subset of seats), a difference between the occupancy probability and a reference probability be determined, and that the seats be ranked in the testing sequence based on this difference. For example, a value of 50% can be specified as the reference probability. Seats that are 50% occupied are therefore tested first. Seats that are 90% occupied, as well as seats with a 10% occupancy probability, are tested later because for these seats, a reliable assumption about seat occupancy exists, and therefore knowledge of the actual seat occupancy is less important.

[0011] Furthermore, it is preferred that for each seat (or at least for some of the seats) an average information content of the knowledge of the seat occupancy is determined and that the seats are ranked in the test sequence according to this average information content. This procedure is similar to the difference method described above, whereby in this embodiment a particularly precise determination of the test tire sequence is possible by evaluating the average information content. This allows the relevant information to be determined and made available early in the evaluation. This procedure can be particularly advantageous if the result of the evaluation can assume more than two values ​​(e.g. unoccupied, occupied by a baby seat, occupied by an adult).

[0012] With the described method, preferably all seats of the motor vehicle are tested, with the exception of the driver's seat. Alternatively, it is preferred that all seats of the motor vehicle are tested. As a further alternative, it is preferred that only some of the seats of the motor vehicle are tested.

[0013] The test sequence is preferably fixed. For example, the test sequence could be based on a presumed probability of occupancy and a deviation for the driver's seat: front passenger seat, seat behind the front passenger seat, seat behind the driver's seat, seat in the middle of a rear bench, driver's seat. The seat most likely to be occupied is tested first (which is not the driver's seat) so that, for example, a warning signal can be issued for this seat first if the seat belt is not fastened. When determining the test sequence, it is preferable to take into account whether the driver is sitting on the left or right side of the vehicle (i.e. whether the vehicle is a left-hand or right-hand drive vehicle).

[0014] Furthermore, it is preferred that the test sequence be adapted to information from which seat occupancy can be deduced. Any information available in a motor vehicle's control unit can be used and interpreted for this purpose. The test sequence can also be changed dynamically. This means that after a test has already been performed according to step b), a different test sequence can be specified for a subsequent test.

[0015] After the test sequence has been determined in step a), in step b), it is preferably determined which seat is occupied using the test sequence created in step a). To do this, the seat is preferably checked sequentially (i.e., one seat at a time) to determine whether a seat is occupied. The test in step b) is preferably performed using sensors in the vehicle.

[0016] In a preferred embodiment of the method, measurement data are obtained before step b) and are evaluated to check the occupancy in step b).

[0017] In a further preferred embodiment, steps a) and b) are carried out periodically one after the other, with the seat to be checked next depending on the result of the previous occupancy. This takes dependencies in the occupancy probability into account in a particularly advantageous manner. For example, the seat on the back bench to be evaluated first can depend on the evaluation result of the front passenger seat. If the front passenger seat is occupied, then the seat behind the driver, for example, can be considered to be occupied with a particularly high probability. This seat is therefore preferably evaluated particularly early. If the front passenger seat is not occupied, then the seat behind the front passenger seat, for example, can be considered to be occupied with a particularly high probability. This seat is therefore preferably evaluated particularly early.This design allows for passengers, such as children, to be supervised without disturbing the driver when a passenger is present. In the absence of a passenger, the passenger can preferentially sit behind the passenger seat, for example, to be able to communicate more easily with the driver. Therefore, it can be advantageous to dynamically design the test sequence depending on the already known occupancy.

[0018] Preferably, signals from sensors within the motor vehicle are recorded as measurement data. This can be, for example, measurement data from the following sensors: an infrared sensor, a thermal imaging camera, an optical camera, a light barrier, a weight sensor (particularly within a seat), a sensor for (three-dimensional) depth measurement, a (three-dimensional) laser scanner and / or a sound detector. A depth measurement can be carried out in particular using stereo camera(s), a time-of-flight sensor, a lidar scanner, a radar sensor and / or an ultrasonic sensor. Cameras used as sensors can be sensitive to visible light and / or to light invisible to humans, in particular to infrared light. In particular, the interior of the motor vehicle can be illuminated with infrared light for this purpose, without the occupants noticing.

[0019] For example, weight sensors can be installed in the seats. If the mass measured for a seat exceeds a specified minimum value, the seat is preferentially assumed to be occupied. Furthermore, such weight sensors can, for example, differentiate whether an adult or a child is sitting in the corresponding seat. If a child is detected in this way, the deployment of an airbag for that seat can be suppressed in the event of an accident. This can be advantageous because, in certain accident situations, an airbag can endanger rather than protect a child.

[0020] In a further preferred embodiment of the method, the measurement data comprise at least one image from an interior camera.

[0021] The interior camera is preferably a CCD camera or a CMOS camera. In addition to a monocamera measurement, depth measurement is also possible additionally or alternatively with the interior camera, for example by using an interior camera designed as a stereo camera or as a time-of-flight camera. In order to save computing power, the interior camera is preferably configured to record black and white images. Color information is not always required and could unnecessarily slow down analysis. Alternatively, color and / or infrared information can be evaluated for more precise determination, which can enable even more precise detection, classification, and / or identification of seat occupancy. The interior camera is preferably arranged and aligned in such a way that the occupancy of every seat in the motor vehicle can be detected.This allows the process to be carried out with a single interior camera, saving costs and space for additional sensors.

[0022] In a further preferred embodiment of the method, in step b) a section of the image from the interior camera is evaluated for the sequential testing of each individual seat.

[0023] The image from the interior camera is preferably divided into a plurality of sections, with at least one section being assigned to each seat. It is not necessary for the entire image from the interior camera to be used. Therefore, there may also be areas of the interior camera image that are not assigned to a seat and that are preferably not considered in an evaluation. Sections may also be used in multiple evaluations, which is particularly advantageous when seats are partially obscured. Sections may also overlap, making them ideal for evaluating a specific seat. This can reduce the computing power required.

[0024] Preferably, the sections are selected such that an occupant on a seat, in a normal seating position, at least partially and preferably completely fills the corresponding section. If it is detected that an object at least partially fills a section of the image from the interior camera, the corresponding seat is preferably assumed to be occupied. The image from the interior camera is preferably evaluated using appropriate software for this purpose.

[0025] Preferably, the interior camera is used to capture and analyze individual still images. A still image can be analyzed with less computing power than a moving image (i.e., a sequence of several still images).

[0026] Furthermore, it is preferred that the shape of an object is analyzed in a section. This can be used to differentiate, for example, whether a person or an object is sitting on a seat. If such an object is detected, the deployment of an airbag can be suppressed, for example in the event of an accident. If an airbag hits an object, occupants of the motor vehicle can be endangered. It is also preferred that, in particular for such a distinction between a person and an object, a movement of the object is analyzed. If the detected object moves, this can indicate that a person is sitting on the corresponding seat. If, on the other hand, the detected object does not move, this can indicate an object. By detecting the movement of objects, it is also possible to detect, for example, when an object is being loaded into the motor vehicle or when a person gets into the motor vehicle.The detection of object movements can be achieved, in particular, by comparing individual still images from the interior camera. To analyze a movement, the interior camera preferably records and compares several (preferably 3 to 8) still images for a single test at specified intervals (preferably in the range of 0.1 to 2 seconds).

[0027] Preferably, several sections of the interior camera image are assigned to each seat. This makes it possible, for example, to differentiate whether an adult or a child is sitting in the corresponding seat.

[0028] In a further preferred embodiment, classification is performed using pattern recognition within one or more sections of a camera image. This allows for a particularly robust classification of the occupancy and the occupant object. For example, a distinction can be made between an adult, a child, a baby seat, and an object.

[0029] Preferably, the sections of the image from the interior camera are fixed. Alternatively, it is preferred that the sections be adapted to additional information that may indicate seat occupancy. For example, if a minimum mass is measured on a seat using a weight sensor, this may indicate that a person is sitting on that seat. In such a case, the interior camera can differentiate whether there is a person on the seat or just a (sufficiently heavy) object. For the corresponding seat, for example, a larger section of the image from the interior camera can be checked so that the shape of the person or object can be analyzed.

[0030] In a further preferred embodiment of the method, the test sequence is stored in a control unit.

[0031] The control unit preferably comprises at least one memory in which the test sequence is digitally stored. The test sequence can be fixed or changeable. For example, the test sequence can be changed by manual setting, e.g., via an on-board computer or a diagnostic device. The test sequence can also be changed automatically using a predefined algorithm. For example, the test sequence can be adjusted with the algorithm using the measurement data described above. If this measurement data indicates, for example, that a seat might be occupied, this seat will be checked preferentially over other seats. For this purpose, the control unit preferably has software with which, in particular, the measurement data described above can be evaluated.Furthermore, it may be advantageous to consider other requirements, such as statutory maximum durations for the evaluation of a specific seat. These can increase the value of the information regarding occupancy and / or classification of the occupying property and are therefore preferentially considered when establishing the test sequence.

[0032] Furthermore, it is preferred that the control unit has at least one piece of software designed and configured to analyze the image from the interior camera. This means, in particular, that the control unit can preferably detect seat occupancy from the image from the interior camera.

[0033] In a further preferred embodiment of the method, before step a) it is checked for each seat whether an associated belt is closed, wherein seats with a closed belt are checked last in step b).

[0034] The belt buckles preferably have sensors that can detect whether a belt is fastened or not. If a belt is detected as fastened, the corresponding seat can be considered to be very likely occupied. This seat can be checked for occupancy with a lower priority than seats with an unfastened belt. As previously described, it is preferable not to assign seats with a very high probability of occupancy to the test sequence according to occupancy probability, but rather to place them at the end of the test sequence, contrary to the arrangement of the seats in descending order of occupancy probability. This applies in particular to seats where a belt is detected as fastened. In particular, if a belt is already fastened, no warning signal is required to indicate that an occupant is not buckled up.

[0035] In a further preferred embodiment of the method, in step a) an occupancy probability is adjusted based on statistical usage data from previous uses of the motor vehicle.

[0036] In this embodiment, it is preferably recorded which seats of the motor vehicle are occupied at which point in time. For this purpose, for example, the proportion of the total operating time or the duration of driving cycles of the motor vehicle that this seat was occupied can be stored for each seat (in particular in the control unit). A proportion stored in this way can be used in step a) as the occupancy probability of the seat. Furthermore, it is preferred that the seat occupancy (and in particular the corresponding proportion of the total operating time of the motor vehicle) be stored as a function of further parameters. Possible parameters of this type include all variables that can be determined using sensors of the motor vehicle and on which the use of the motor vehicle may depend.For example, the occupancy probability for each seat can be determined depending on the time of day, the season, and / or the weather conditions (which can be detected via a temperature sensor and / or a rain sensor and / or weather maps). This makes it possible to take into account, for example, that children may be more likely to sit in the rear seats in bad weather and / or in winter. The time of day can also be taken into account. For example, it may be that children sit in the rear seats particularly at certain times of the day. Preference is given to vehicle usage habits (particularly through the use of statistical usage data).

[0037] In a further preferred embodiment of the method, in step a) at least one of the following variables is used to adjust the occupancy probability: Information about the use of seat heating, Information about adjusting a seat position, Information on the use of controls in the motor vehicle located near the seat, Information on the use of a window regulator, Information about the position of noise sources, Information about opening and / or closing a door.

[0038] The information mentioned may indicate that a seat is occupied. The seat heating and the seat position adjustment are usually used when the corresponding seat is occupied. If a control element that is (only) accessible from a seat is operated, this may indicate that the corresponding seat is occupied. The fact that a control element is located near a seat is to be understood in particular to mean that an occupant of the corresponding seat can operate the control element or that the control element is designed and intended to be operated by the occupant of that seat. If a reading lamp is switched on, for example, this may indicate that a specific seat is occupied. It may also indicate that a seat is occupied if a button to operate the window lift is pressed.If noises are detected in the motor vehicle (e.g., via a microphone), this may indicate that a person is present at the corresponding location from whom the noise is emanating. Noises can be localized using multiple microphones. The opening and / or closing of the motor vehicle's doors is preferably monitored electronically. If a door is opened, this may indicate a change in the occupancy of the corresponding seat. As long as a door is not opened, the occupancy of the corresponding seat can be assumed to be unchanged.

[0039] If there are indications that a seat may be occupied, that seat will be given priority during the check. This means that the corresponding seat is given priority in step a) and a higher probability of being occupied.

[0040] In a further preferred embodiment of the method, an initial seat occupancy is detected immediately after the motor vehicle is started, wherein a periodic check is carried out during the operation of the motor vehicle to determine whether a change in the seat occupancy has occurred.

[0041] The initial seat occupancy is the seat occupancy that exists immediately after the motor vehicle has been started, i.e. in particular immediately after an internal combustion engine or another drive motor of the motor vehicle has been switched on. The detection of the initial seat occupancy is preferably completed as quickly as possible after the motor vehicle has been started. This means that a warning signal can be issued as quickly as possible after start-up if an occupant is not wearing a seat belt. Such a warning signal is preferably issued no later than one second after the motor vehicle has been started so that the occupant in question can, for example, fasten their seat belt before the motor vehicle starts moving or before the motor vehicle reaches a (specifiable) minimum speed. The warning signal is preferably not issued before the minimum speed has been reached for the first time after the motor vehicle has been started.The initial seat occupancy check preferably includes performing steps a) and b) at least once.

[0042] It can be assumed that the seat occupancy is unlikely to change during operation of the motor vehicle. Therefore, preferably after the initial seat occupancy has been detected, a periodic check is carried out to determine whether a change in the seat occupancy has occurred. Preferably, a seat occupancy check is carried out after a specified time interval. Such a check preferably comprises at least the implementation of steps a) and b). In step a), a seat occupancy detected during a previous check is preferably taken into account. In particular, the occupancy probability of a seat can be assumed to be particularly high if this seat was detected as occupied during the previous check.

[0043] The time interval between successive tests preferably has a length in the range of 1 to 20 seconds, in particular in the range of 5 to 10 seconds. It is preferred that the length of the time interval is selected as a function of at least one of the following parameters or events: a speed of the motor vehicle, a duration of a driving state, an opening or closing of a belt, an opening or closing of a door, a change in a signal from a weight sensor in a seat, an actuation of a control element, a switching on or off of a seat heater, an adjustment of a seat position and / or an actuation of a window lift. If the motor vehicle is moving at high speed over a longer period of time, for example, it can be assumed that the motor vehicle is traveling on a motorway.In such a driving condition, a longer time interval can be selected because a change in seat occupancy is unlikely. However, if a door is opened while the vehicle is stationary, a change in seat occupancy can be assumed to be very likely. It is particularly preferred that the time interval corresponds to the duration of a single test, so that after the completion of one test, the next test begins without a waiting period.

[0044] Also to be described here is a computer program which is set up to carry out all steps of the method and a machine-readable storage medium on which this computer program is stored.

[0045] Further details of the invention and an embodiment, to which the invention is not limited, are explained in more detail with reference to the drawing. It shows: Fig. 1: a schematic flow diagram of a method for detecting seat occupancy in a motor vehicle.

[0046] Fig.Figure 1 shows a schematic flow diagram of the described method. After the motor vehicle has been started at step I, the seats of the motor vehicle are first sorted according to step a) in descending order of occupancy. In this example, the test sequence is: driver's seat, front passenger seat, rear seats. Subsequently, according to step b), the occupancy of the seats is checked sequentially in the test sequence. First, in step b), a distinction is made as to whether a belt on the driver's seat is fastened. If this is not the case (path VI), the driver's seat is checked first at step VII. Depending on the result of this check, a warning signal for the driver's seat may be issued at step X. However, if the driver's belt is fastened, the driver's seat is not checked and the test of the front passenger seat begins at step VIII (path V). The test of the rear seats then follows at step IX.A warning signal is also issued for the front passenger seat and the rear seats if necessary at X. Steps a) and b) performed at I after the vehicle has been started together form a check of an initial seat occupancy II.

[0047] After the initial seat occupancy check II, a periodic test III follows. Periodic test III comprises steps a) and b), which are carried out alternately. Step b) shows that the seats are tested in the following order: driver's seat VII, rear seats IX, front passenger seat VIII. After each step b) is carried out, the test sequence is redefined according to step a), so that the representation of step b) is only an example. During periodic test III, a warning signal X may also be given in step b) if a seat is occupied and the corresponding belt is not fastened. Periodic test III is carried out until the motor vehicle is switched off at step IV.

Claims

[1] Method for detecting seat occupancy in a motor vehicle, comprising at least the following steps: (a) establishing a seat testing sequence using at least a seat occupancy probability, b) Sequential check of seat occupancy using the check sequence established in step a). [2] Method according to claim 1, wherein measurement data are obtained before step b) and are evaluated to check the occupancy in step b). [3] Method according to claim 2, wherein the measurement data comprise at least one image from an interior camera. [4] Method according to claim 3, wherein in step b) a section of the image of the interior camera is evaluated for the sequential testing of each individual seat. [5] Method according to one of the preceding claims, wherein the test sequence is stored in a control unit. [6] Method according to one of the preceding claims, wherein before step a) it is checked for each seat whether an associated belt is fastened, and wherein seats with a fastened belt are checked last in step b). [7] Method according to one of the preceding claims, wherein in step a) an occupancy probability is adjusted based on statistical usage data of previous uses of the motor vehicle. [8] Method according to one of the preceding claims, wherein in step a) at least one of the following variables is used to adjust the occupancy probability: - Information on the use of seat heating, - Information about adjusting a seat position, - Information on the use of controls in the motor vehicle located near the seat, - Information on the use of a window regulator, - Information about the position of noise sources, - Information about opening and / or closing a door. [9] Method according to one of the preceding claims, wherein an initial seat occupancy is detected immediately after the motor vehicle has been started, and wherein a periodic check is carried out during operation of the motor vehicle to determine whether a change in the seat occupancy has occurred. [10] Control device for a motor vehicle, which is designed to carry out the method according to one of the preceding claims. [11] Computer program which is arranged to carry out all steps of the method according to one of claims 1 to 9. [12] A machine-readable storage medium on which the computer program according to claim 11 is stored.

Images