EP3102463A1

EP3102463A1 - Method and device for generating a signal representing an occupation of a vehicle seat, corresponding computer program, and machine-readable storage medium

Info

Publication number: EP3102463A1
Application number: EP15704283.9A
Authority: EP
Inventors: Thomas Lich; Joerg Moennich; Gian Antonio D'addetta
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2014-02-06
Filing date: 2015-02-06
Publication date: 2016-12-14
Also published as: US20170166086A1; WO2015118127A1; CN105980216B; DE102014202130A1; CN105980216A

Abstract

The invention relates to a method for generating a signal representing an occupation of a vehicle seat, the method comprising the steps: detecting a signal of a sensor system for detecting a dynamic characteristic quantity with respect to the vehicle seat; evaluating the signal, in particular a progression of the signal; generating a seat occupancy signal as a function of the evaluation of the signal, wherein the seat occupancy signal is suited to indicate if the vehicle seat is occupied by a person or an object.

Description

description

title

Method and device for generating a signal representing occupancy of a vehicle seat, and a corresponding computer program and machine-readable storage medium

State of the art

Since the introduction of the legal seat belt obligation in Germany in 1977 and the introduction of the seatbelt in the 1970s, the number of those killed has been reduced significantly from 21,000 to less than 4,000. In addition to other restraint systems, the seat belt provides the greatest contribution to protecting the occupant in the event of an accident, with almost 75% - 80% of the restraint.

The possibilities for manipulating the current systems are arbitrary. It is, for example, of concern that more and more car owners at their own expense as well as their own risk and to the exclusion of manufacturer responsibility leave the corresponding belt reminder in specialist workshops on the software side.

The seatbelt application rate and its relevance to the percentage of unbelted, killed car occupants from 2010 illustrate the above mentioned. Connection still. It becomes clear that in the Federal Republic of Germany in 2010 every fifth inmate killed is due to non-buckling.

Special relevance of this topic also comes to the US. In response to a petition of August 2013 [Th. 2013-21 128 and 2013-0095], the NHTSA mentions that about 200 lives a year are saved by road safety in traffic accidents. It was found that the belt buckling rate in the US in 2012 was about 86%, yet the remaining 14% are unglued and thus potentially at risk of fatal accident in a collision. Usually, it is checked on the basis of seat occupancy recognition devices, if the seat belt buckle is plugged, ie, whether the preferably metallic tongue of the belt system is engaged in the buckle. Based on this, the driver is usually warned by acoustic, optical or visual signals that one or more occupants of his vehicle are not properly strapped or not at all buckled. However, a manipulation of the system is still possible, for example, by guiding the belt behind the occupied seat or behind the back of the occupied seat and plugging in the buckle or by the use of dummy belt tongues. It can therefore be concluded that inmates of European vehicles or vehicles that do not buckle up for the European market can not benefit from the significantly higher level of vehicle safety. Neither can any crash loads be reduced by the first used in 1985, irreversible, pyrotechnic belt tensioners, nor benefit these inmates from the ever-expanding Pre-crash systems.

The need for a sustainable "motivation" of this so-called "Gurtmuffel" to put on the seat belt is obvious for moral and consumer protection reasons. Experience in other European countries shows that even with a significant increase in fines a "saturation level" is achieved and remains a residual inmates in the low double-digit or high single-digit percentage range remains that does not buckle up Possibilities are used to achieve the highest possible level of vehicle occupant.

One way to unambiguously identify a correct and correct Angurtvorgang consists in the utilization of the belt extension length. Alternatively, optical methods could be used to detect the Gurtanschnallzustandes, which could be used, for example, to determine the occupant position. However, the current state of development in optical methods of interior sensor technology so far still no optical detection, whether a driver is correct or even strapped. Problem of all methods described above is the fact that the real sitting position in the field is not or only less taken into account. Thus, a proper detection of Gurtanschnallstatus from the information of the belt extension length is difficult. Overall, all mentioned and known systems in common that so far for a Belt-Reminder warning and / or locking function neither the belt pullout information nor a belt force information is used.

German patent application no. 10 2008 042399 discloses a system which, based on an interior sensor system, e.g. a Gurtauszugslängen- information of the Gurtautomaten, the buckle switch and a seat occupancy recognition and a power-on information of the vehicle detection of a so-called. Belt Misuse state, i. the presetting of a correctly inserted belt performs.

From the German patent application no. 10 2010 029790 a system is known, which recognizes by means of a Winkelmessprinzips at the lower pivot point (Gurtankerpunkt or -Verankerungspunkt between belt system and fixed system such as vehicle structure / body or seat) the seat belt. This principle is based on an angle measurement by a contact sensor or another angle-measuring sensor. The need may arise to perform a measurement by means of a contactless sensor, since other sensor requirements are given due to the available space.

Known variants for "kicking out" a belt warning system with sensor in the belt buckle or pullout detection and seat occupancy detection (seat mat) that is common today are listed below:

1 . Deactivation of the system by, for example, a workshop

2. Use of a "dummy belt tongue" (see FIG. 4)

3. Plug the belt and then sit on the webbing (see Fig. 5)

4. Guide the webbing behind the seatback (see Fig. 6)

5. Insert the strap of an unused seat (see Fig. 7)

6. Use of an extension length bracket (see Fig. 8) The table of Fig. 9 comparatively shows the recognizability of the various systems with respect to the types of belt misuse and refers to the prior art.

Variant 3 "Plugging the belt and then sitting on the webbing" will not be used by many people, as the comfort of sitting significantly suffers.A workshop deactivation in accordance with Variant 1 could be restricted for example by legal regulations (for example, examination at main inspection).

The misuse use cases shown in the table of FIG. 9 can be recognized by known systems. Another use case has not yet been considered:

What happens when you take an item, e.g. a guitar case, a heavy box, etc., placed on the front seats and not buckled?

Disclosure of the invention

The focus of the present application is on the above-mentioned use case, namely the placement of a heavy object on the front seats without performing a seatbelt.

The aim is to ensure that no warning function or seatbelt interlock function is activated even if the seat belt is not fastened, as there is no person in the respective seat. The system (warning function / interlock) should be designed so "comfortable" that actually only in a not or not properly strapped person a warning / inhibition and not in (person-like in the sense of weight or shape) objects a warning / inhibition occurs ,

Therefore, it can be regarded as an object of the present invention to realize the unambiguous recognition of a seat occupied only by an object and not by a human, and thereafter, no interruption of the engine performance (inhibition). It can thus be regarded as a further object of the invention, for the underlying use case, to make a decision as to whether a person or a non-human (for example object) is sitting on a seat, without using a belt sensor system.

This object is achieved by a method and a device for generating a signal representing an occupancy of a vehicle seat and a corresponding computer program and machine-readable storage medium according to the independent claims.

The determination of the object is based on the dynamic

Characteristics of seat-integrated or seat-near sensors. It can the

Sensors, for example, be a weight sensor based on absolute weight measurement, but also be a seat mats-based solution or based on a capacity measurement of built-in seat or headliner sensors.

Also conceivable are sensors that are installed in the headrest or center armrest.

The basis of the method is the evaluation of the respective characteristics of the seat-integrated or seat-near sensors. The parameters are typically measured force-time profiles, but also current-time profiles depending on which type of sensor is used. In this case, the different characteristics of a person sitting in relation to an object placed on the seat are taken up. This characteristic differs both in the single-seat operation, i. directly at the state transition from "seat unoccupied" to "seat occupied", as well as in the first driving state, e.g. From the approach to the speed of a "throttling process" (for example, due to a seatbelt interlock system in human recognition.) Significant factors influencing this characteristic are the different inertial effects even with the same mass, but also that due to the active muscle tension and inhomogeneous weight distribution - a human being a clearly changed pattern is conditional.

Advantageous embodiments of the invention will become apparent from the dependent claims and the following description of exemplary embodiments.

Brief description of the drawings Fig. 1: Real traffic accident, fatal accident consequences due to belt misuse (belt support)

Fig. 2: Exemplary Gurtfehlbenutzung, belt anchoring by connected to the B-pillar & sill bar

Fig. 3: Exemplary Gurtfehlbenutzung, belt anchorage connected by the B-pillar swivel

Fig. 4: "Dummy belt tongue" for deactivating the seat belt buckle (belt misuse) variant 2

Fig. 5: "Sitting on the belt" variant 3

Fig. 6: Belt guide behind the seat back variant 4

Fig. 7: Inserting an unused belt (passenger) variant 5

Fig. 8: "belt extraction clamp" variant 6

Figure 9: Recognizability of the various systems with respect to the types of Gurtfehlbenutzung and previously registered or published principles.

Fig.10: pattern comparison of different seat occupancy

Fig.1 1: Block diagram of a first simple embodiment

Fig.12: Block diagram - Coupling of different sensor information

Embodiments of the invention

In a first implementation variant, the detection is carried out whether an object or a person is on the seat by evaluating the Einsitzvorgangs, eg before the start of the ignition, but after opening the vehicle or a vehicle door. The value supplied by the seat-integrated or seat-near sensor system during initialization, ie after power-on, is for example 1000. All values shown below refer to this reference value. The initialization is no longer done in modern vehicles after the ignition, but it is already with the opening of the vehicle (eg. By a remote control to open the vehicle) a (partial) current supply. For a possible first realization results, under the condition that an energization of the seat-integrated or seat-near sensor can be done and this provides a force, current or other value, the following relationship:

In a single-seated operation, each occupant experiences a typical pattern, which is additionally dependent on the position of the seat relative to the instrument panel and on the person (body proportions). This case differs markedly from the exemplary use case of loading the seat by an object, cf. FIG. 10.

The schematic measurements shown in FIG. 10 show a differentiation possibility by a simple threshold comparison. A possible evaluation provides to form the difference to the initial value, to calculate the absolute value of it and to query about a threshold value. Each time the threshold is exceeded, a counter is incremented, which is then also queried via a further threshold value and a corresponding one

Status flag generated. In the example, the threshold value 3 was set here. As can be seen, an immediate evaluation is not possible, but requires the presence of a certain number of successive measurement points. However, the differentiation is clear, so that a differentiation between the use case "object" and "human" can be safely carried out. The influence of different scanning variations, with correspondingly lower sampling, can be countered by an additional weighting.

It should be noted that a simple threshold query is not sufficient. The separation around the two values must be considered as measurement noise and tolerance. Therefore, a dynamic evaluation is required, which allows observation of the relative size.

A flow chart for a variant of the method of the present invention is shown in FIG. 11. In this case, no vehicle dynamics information such as speed or vehicle longitudinal deceleration is used. As a result, the Information is used as an input value for checking whether a correct fastening process is present, ie if no object was detected on the seat, but a person was recognized, it must be checked whether it is correctly fastened. If an item has been detected, it does not.

For this purpose, in step 101 differences are first formed between a predetermined initial value and the sensor signal of a seat-integrated sensor system. These are passed to step 102 as relative changes. In step 102, an amount is then formed of the difference formed.

The amount is compared against a first predetermined threshold in step 103. In an embodiment variant, the first predetermined threshold value can be changed. In the flowchart, this is indicated by the incoming arrow labeled "Thd / Threshold." Each time the

Threshold of the amount has been exceeded, carried out in step 104, an increase of a counter.

The counter reading is compared in step 105 with a second predetermined threshold value. In one embodiment, the second predetermined

Threshold to be changed. In the flow chart, this is represented by the incoming arrow labeled "Threshold Z".

In step 106, depending on the threshold comparison in step 105, the generation of a signal which is suitable for indicating whether the vehicle seat is occupied by a person or an object takes place. This signal can be transmitted via a suitable communication interface, for example. Via a CAN bus to other components in the vehicle. For example. to a system that checks whether a belt system for the vehicle seat has been correctly created when it has been detected that the vehicle seat is occupied by a person.

A supplementary design provides for additional consideration of vehicle dynamics information such as speed or acceleration information. The algorithm is only executed if the speed is above or below a defined threshold. It is also conceivable to carry out only situation-adaptive evaluations, in which the internal sassen or objects is to be expected, for example when accelerating and / or braking. This could potentially increase the stability of the solution. For this purpose, in the invention, the vehicle dynamics information already available, for example, on the CAN bus is evaluated as additional input. As base information, the vehicle speed and, if available, the longitudinal acceleration (ax) is used. This enables dynamic observation.

A possible addition uses the driving dynamics sensor and includes in one embodiment driving dynamics variables such as speed or

Acceleration in the longitudinal direction. An extension by driving dynamics variables such as accelerations in all three spatial directions, rotation rates in all three directions, wheel speeds, brake pressure or braking pressures of the individual wheels, steering wheel or steering angle information on the wheel, brake and accelerator pedal position is also conceivable, thereby increasing the Auswerterobustheit.

Another implementation variant is based on saving the last state / status of the sensor information when the vehicle is switched off. It would be possible, e.g. the storage of a value or a value profile in the appropriate control unit. When the vehicle is restarted, the dynamic information is compared with the stored values or value profiles according to the method described above. From this it can be concluded that no change is a "rigid" charge / object, as a human inmate would always behave differently and not remain static over a longer period of time.

A general flow chart taking into account different sensor information is shown in FIG. 12. By way of example, the sensor for static and dynamic measurement can be image information with subsequent object recognition (eg interior camera), a thermal profile, a noise profile, weight information (eg seat-integrated sensors). The sensor system for detecting an external excitation includes the measurement of a reaction of an object or person to an external excitation, for example "starting" by seat kinematics and / or driving dynamics and / or vehicle dynamics (braking / accelerating). In addition to vehicle dynamics information, other measured variables or states provided for other functions can also be used for plausibility checking or threshold value adaptation. For example, microswitches in the door handle can detect if the door has been opened / closed from inside or outside. Operation of the seat adjustment with the door closed and the driver-side seat side can also be closed to a person sitting on the seat

Detection may be done in a controller, such as e.g. the

Airbag control unit. Other control devices with corresponding communication possibilities, such as A CAN bus is also conceivable (e.g.

Door control unit), since these are not time-critical evaluations. Control of other components of the vehicle, such as the starter are also conceivable. As overriding advantage of the invention is the protection of the individual in the

Road traffic, since a typical application of a proper operation by loading the seat with an object is distinguished from a more typical use case of a seated vehicle occupant. A further advantage is that the functionality "allows the seat to be fastened on the seat without him", otherwise the object must always be fastened in a mandatory manner, a lack of which would have far-reaching consequences for the acceptance of such a device, especially in the USA Further advantages over the prior art:

• In the event of an accident, when charge detection (ie, no human), the ignition of unnecessary retention means can be suppressed, for example, to prevent repair costs. Accidental acceleration of the charge (for example, by the deployment of the airbag) can thus also be avoided.

• In the event of an accident, appropriate means of restraint can be reparameterized before cargo loading, especially for the purpose of cargo protection / protection of the occupants, or to prevent them from flying around (eg modified airbag deployment to occupants).

· The device represents a cost-effective alternative to indoor cameras.

Claims

claims

1 . Method (100) for generating a signal that has an occupancy of a

Vehicle seat of a vehicle represented, with the steps:

Detecting a signal of a sensor for detecting a dynamic characteristic with respect to the vehicle seat;

Evaluating (101 - 106) the signal, in particular a course of the signal;

Generating a seat occupancy signal depending on the evaluation of the signal, wherein the seat occupancy signal is adapted to indicate whether the vehicle seat is occupied by a person or an object.

2. Method (100) according to claim 1, wherein the sensor system is suitable for force measurement and the parameter represents a force-time curve and / or a current-time curve.

The method (100) of claim 1 or 2, wherein the method is performed only when the speed of the vehicle falls below or exceeds a predetermined speed threshold.

4. The method according to claim 1, wherein the course of the parameter is evaluated during the seated operation, in particular from the period after the opening of the vehicle and before the start of the ignition of the vehicle.

5. The method (100) according to any one of claims 1 to 3, wherein the course of the characteristic during a first driving state, in particular from a drive of the vehicle up to a speed of a throttle process of the vehicle, is evaluated.

6. Method n (100) according to any one of claims 1 to 3, wherein the method is carried out only if with a high dynamics of the vehicle seat occupying subject or person is to be expected, in particular when accelerating and / or braking the vehicle.

The method of claim 1, wherein in the step of evaluating, differences of the detected signal are formed into a predetermined initial value and a counter is incremented each time the absolute value of the difference exceeds a predetermined first threshold value Generating a seat occupancy signal indicating that the vehicle seat is occupied by a person is generated when the counter exceeds a second predetermined threshold.

8. The method (100) according to any one of the preceding claims, wherein in the step of evaluating the difference takes place weighted.

9. Method (100) according to claim 1, wherein in the step of evaluating an adjustment is carried out with a value and / or value profile, wherein the value or the value profile represents a state and / or status of the evaluation during a preceding shutdown of the vehicle ,

10. The method (100) according to one of the preceding claims, wherein in the step of generating as the seat occupancy signal, a status flag is set.

1 1. Computer program which is set up to carry out all steps of the method (100) according to one of the preceding claims.

12. Machine-readable storage medium on which the computer program according to claim 1 1 is stored.

13. Device, in particular electronic control unit, which is configured to carry out all steps of the method (100) according to claim 1 to 10.