DE102011123092B4 - Method for controlling a closure element arrangement of a motor vehicle - Google Patents

Abstract

Method for controlling a closure element arrangement (1) of a motor vehicle, wherein the closure element arrangement (1) has a closure element (2), a drive arrangement (3) assigned to the closure element (2), a control arrangement (4) and a sensor arrangement (5) with at least one sensor element (6, 7), wherein, as part of an operating event monitoring, sensor events in the sensor measurement values (8, 9) of the sensor arrangement (5) are monitored by means of the control arrangement (4) to determine whether a predetermined operating event is present, and the drive arrangement (3) is controlled depending on the result of the operating event monitoring, wherein the operating event monitoring is based on at least one operating event model stored in the control arrangement (4) and wherein the operating event model is generated and/or parameterized in normal operation, characterized in that, in the event that a sensor event was not recognized as an operating event as part of the operating event monitoring and that a control of the drive arrangement (3) was subsequently triggered manually, an operating event model is generated or parameterized from the sensor event by means of the control arrangement (4). such that the sensor event is recognized as an operating event during further operation.

Description

The invention relates to a method for controlling a closure element arrangement of a motor vehicle according to the preamble of claim 1 and a closure element arrangement of a motor vehicle according to the preamble of claim 13.

Today's motor vehicles are increasingly equipped with motor-operated locking elements. Such locking elements can be, for example, doors, in particular sliding doors, flaps, in particular tailgates, rear lids, engine hoods, loading floor or similar of a motor vehicle. In this respect, the term "locking element" is to be understood broadly in this case.

A convenience function that is becoming increasingly important today is the automatic operation of the motorized tailgate of a vehicle. EN 10 2009 019 673 A1 relates to a motor vehicle with a flap which can be opened or closed automatically following a contactless command by a person. A sensor is provided which forms at least two effective areas, each of which has a horizontal area extending from the motor vehicle for detecting the command.

With the well-known comfort function ( EN 20 2005 020 140 U1 ), from which the invention is based, it is provided that an operator-side operating event, in this case an operator-side foot movement, causes the motorized opening of the tailgate. For this purpose, a drive arrangement assigned to the tailgate, a control arrangement and a sensor arrangement are provided. The control arrangement monitors the sensor events of the sensor arrangement that occur in the sensor measurement values to determine whether the above operating event has occurred. Depending on the result of this operator event monitoring, the drive arrangement is controlled accordingly.

The above known method for sensory detection of an operating event enables particularly convenient motorized operation of the tailgate of a motor vehicle. The design of the control arrangement and the sensor arrangement are crucial for the reliability of the operation. Difficulties can arise in particular if different operators of the motor vehicle implement the foot movement in question in very different ways, particularly due to different anatomical conditions. This can lead to an operator movement not being recognized as an operating event, even though the operator subjectively performed this movement correctly.

The invention is based on the problem of designing and developing the known method in such a way that a high level of reliability in operation is ensured even when there are several potential operators.

The above problem is solved in a method according to the preamble of claim 1 by the features of the characterizing part of claim 1.

The fundamental consideration is that part of the operator event monitoring can be changed during normal operation. In this case, “normal operation” means the state of use in which the motor vehicle is available to the user for normal operation. Setup or installation work carried out by the manufacturer during production of the motor vehicle before it is delivered cannot be assigned to normal operation and takes place before normal operation. The flexibility gained in operator event monitoring enables operator event monitoring to be adapted to the operator who is active in the operation of the motor vehicle.

In detail, it is proposed that the operator event monitoring is based on at least one operator event model stored in the control arrangement and that the operator event model is generated and/or parameterized in normal operation.

This basically makes it possible to teach in operating events from scratch. On the other hand, it is also conceivable that predefined operating events can be parameterized to the movement sequences of the respective operator in normal operation.

The operator event model is a data model that is directed at the sensor measurement values of the sensor arrangement and contains a set of variables that are characteristic of the respective operator event. The operator event model can also include entire courses of the sensor measurement values of the sensor arrangement. Other variants for the design of the operator event model are conceivable.

Of particular importance for the proposed solution is the fact that the creation and parameterization of the operator event model can be carried out during normal operation. This makes the ongoing Adaptation of the respective operator event model possible during operation.

The particularly preferred embodiments according to claims 2 to 4 relate to the use of the motor vehicle by at least two potential operators as required. Optimum user-friendliness is achieved according to claim 4, according to which an operating event model is assigned to each potential operator. This ensures that an operating event is always reliably recognized for several potential operators, even if the potential operators have very different movement behavior.

In the further preferred embodiments according to claims 5 and 6, the operator event model is generated or modified by putting the control arrangement into a learning mode. This is a variant in which the operator retains full control over a change in the operator event model, which is seen as an advantage depending on the operator.

The situation is different in the particularly preferred embodiments according to claims 7 and 8. Here, the control arrangement itself recognizes, according to different strategies, that the operator event model needs to be generated or parameterized. These automatic functions are not noticeable to the operator and lead to a particularly high level of operating comfort.

The proposed solution has proven particularly effective when the operating event is a foot movement of the operator according to claims 10 and 11. Experience has shown that the movement sequences of different operators exhibit very pronounced differences here.

According to a further teaching according to claim 13, which also has independent significance, a closure element arrangement is claimed which operates according to the proposed method. Reference may be made to all embodiments which are suitable for describing the closure element arrangement as such.

• 1 den Heckbereich eines Kraftfahrzeugs mit einer vorschlagsgemäßen Verschlusselementanordnung zur Durchführung des vorschlagsgemäßen Verfahrens,
• 2 eine Sequenz eines vorschlagsgemäßen, zu erfassenden Bedienereignisses in den fünf Stufen a) bis e) und
• 3 zwei beispielhafte Signalverläufe zweier Sensorelemente der Verschlusselementanordnung gemäß 1 während eines Bedienereignisses gemäß 2.

In the following, the invention is explained in more detail with reference to a drawing which merely represents an exemplary embodiment. In the drawing,

• 1 the rear area of a motor vehicle with a proposed closure element arrangement for carrying out the proposed method,
• 2 a sequence of a proposed operating event to be recorded in the five stages a) to e) and
• 3 two exemplary signal curves of two sensor elements of the closure element arrangement according to 1 during an operating event according to 2 .

The proposed method is explained below using a closure element arrangement 1 of a motor vehicle, which has a closure element 2 designed as a tailgate.

The design of the closure element 2 as a tailgate of a motor vehicle is preferred in the present case. However, with regard to the broad understanding of the term "closure element", reference may be made to the introductory part of the description. In this respect, all statements regarding a tailgate 2 apply accordingly to all other types of closure elements.

The tailgate 2 is assigned a drive arrangement 3, by means of which a motorized adjustment of the tailgate 2 between the 1 closed position shown in solid line and the 1 open position shown in dashed line.

A control arrangement 4 is also provided, which serves, among other things, to control the drive arrangement 3. A sensor arrangement 5 with at least one sensor element 6, 7, which is designed in particular as a proximity sensor, communicates with the control arrangement 4. The sensor arrangement 5, which will be explained later, is designed in such a way that operator movements can be detected by sensors.

In the illustrated and in this respect preferred embodiment, the control arrangement 4 has a central hardware structure. However, it is also conceivable that the control arrangement 4 has a decentralized hardware structure. In this case, part of the control hardware is preferably housed in the sensor arrangement 5. This means that the sensor arrangement 5 has its own intelligence, which can be used for signal preprocessing, for example.

Certain operator movements are defined here as operating events, each of which is intended to trigger a specific reaction of the control arrangement 4. Preferably, the operating event is a foot movement of the operator, as will be explained below.

In order to be able to react to the occurrence of operating events as mentioned above, an operator event monitoring is provided, which is carried out by the control arrangement 4. As part of the operator event monitoring, the control arrangement 4 monitors the sensor events occurring in the sensor measurement values 8, 9 of the sensor arrangement 5 to determine whether or not a predetermined operator event has occurred. The sensor events can be any deviations of the sensor measurement values 8, 9 from the idle sensor measurement values. The idle sensor measurement values 8, 9 are present when there is no external influence on the sensor arrangement 5 by an operator or the like.

Depending on the result of the operator event monitoring, the drive arrangement 3 is controlled accordingly, for example the motorized opening of the tailgate 2.

2 Figures a) to e) show the stages of the sequence of an operating event that includes a back and forth movement of the operator's foot. 3 shows the resulting sensor events in the sensor measured values 8, 9 of the two sensor elements 6, 7, whereby in the representation there the 2 The stations shown a) to e) are marked accordingly in the time domain. This is explained in more detail below.

It is now essential that the operator event monitoring is based on at least one operator event model stored in the control arrangement 4. The operator event model can be any set of variables that is suitable for characterizing the course of the sensor measured values 8, 9 associated with an operator event, so that the occurrence of an operator event in the control arrangement 4 can be recognized by means of a corresponding algorithm.

The above operator event model can be generated and/or parameterized in normal operation as suggested. While the generation of the operator event model usually focuses on recording the history of sensor measured values 8, 9, the parameterization of an operator event model usually involves the assignment of specified model variables.

The generation or parameterization of the operator event model can, for example, be carried out by means of a learning process, which will be explained later. However, it is also conceivable that this takes place automatically, as will also be explained later.

Advantageously, at least two potential operators are provided for the normal use of the motor vehicle, with the control arrangement 4 always determining which of the potential operators is active with regard to the operation of the motor vehicle. Various criteria can be used to determine the active operator. For example, the operator who was the last operator to activate any function of the motor vehicle while the radio key 10 of this operator was in the immediate vicinity of the motor vehicle can be considered active. Preferably, this is the operator who was the last to lock the motor vehicle locking system or the like.

Preferably, therefore, an above radio key 10 serves to identify the operator. Here and preferably, therefore, at least two potential operators of the motor vehicle are each assigned an individual radio key 10, wherein the control arrangement 4 determines from the presence and/or operation of the radio key 10, as explained above, which of the potential operators is currently active with regard to the operation of the motor vehicle.

Because the generation and/or parameterization of at least one operator event model is possible in normal operation, it is easy to store at least two operator event models in the control arrangement 4 that are assigned to different operators. The important thing here is the fact that the operator event monitoring is always based on the operator event model of the currently active operator. This makes it possible to query different movement sequences for different operators within the scope of the operator event monitoring.

In a particularly preferred embodiment, the control arrangement 4 can be put into a learning mode in which an operating event model is generated or parameterized. Within the learning mode, the operator preferably carries out an operating event, wherein the resulting sensor measurement values 8, 9 are used to generate or parameterize the operating event model by means of the control arrangement 4 and then store it, i.e. save it.

When generating the operating event model, it is possible to specify any operator-defined operating events, provided that the associated operator movements can be detected by means of the sensor arrangement 5. For example, in addition to foot movements, other gestures such as "swiping" or "stroking" movements or the like can also be taught as operating events.

In a particularly preferred embodiment, the learning mode is carried out in a user-specific manner, in such a way that the operator event model is stored for a specific operator, in particular for the operator active during the learning.

The learning mode can be set in a variety of ways. For example, the learning mode can be set when a lock button on the remote control key is pressed several times or for a longer period after a predetermined operator action has been triggered, such as switching on the ignition or unlocking.

A further preferred embodiment relates to the case in which a sensor event was recognized as an operating event within the scope of the operating event monitoring. In this preferred embodiment, the active operator is then determined by means of the control arrangement 4, with the operating event model underlying the operating event finally being assigned to the active operator in each case.

In a further preferred embodiment, it can be provided here that all operating event models stored in the control arrangement 4 are always used for operator event monitoring. If one of the operating event models then leads to the detection of an operating event, this operating event model is preferably assigned to the respectively active operator.

What is interesting about the latter variant is the fact that the assignment of the operating event model to the active operator takes place automatically, without the operator having to notice. Such self-optimization of the control arrangement 4 leads to a particularly high level of operating comfort.

Another self-optimization is proposed for the case that a sensor event was not recognized as an operating event within the scope of the operating event monitoring, wherein subsequently, in particular within a predetermined period of time, a control of the drive arrangement 3 was triggered manually, here and preferably via the radio key 10.

The above facts generally allow the conclusion that an operator has attempted unsuccessfully to trigger a function of the locking element arrangement 1, in particular the opening of the tailgate 2, by means of an operating event, in this case by a foot movement, and that after this unsuccessful attempt the operator has switched to triggering the function of the locking element arrangement 1, in this case the control of the drive arrangement 3, manually, in particular via the radio key 10. In such a case, it is preferably provided that the control arrangement 4 generates or parameterizes an operating event model from the respective sensor event (which was not initially recognized as an operating event), in such a way that this sensor event is recognized as an operating event during further operation. It is here and preferably provided that the operating event model generated or parameterized in this way is assigned to an operator, in this case and preferably the active operator. This ensures that the operating event last carried out by the operator will trigger the desired function in the future, even if this failed the last time it was attempted.

Another preferred variant of self-optimization consists in the control arrangement 4 optimizing the respective operating event model practically every time an operating event is detected. For this purpose, it is provided that, by means of the control arrangement 4, every time an operating event is detected, a distance value for the distance between the sensor event on which the detection is based and the respective operating event model is determined and, if necessary, stored. The distance value indicates the extent to which the sensor event on which the detection is based corresponds to the criteria for detecting an operating event specified in the operating event model.

Based on the distance value, a modification of the operator event model is carried out in such a way that a statistical distance value, here and preferably the average distance value, is reduced over a number of the last operator event detections. This means that an operator event that is always only just recognized as an operator event by a certain operator based on the corresponding operator event model leads to a modification of the operator event model. The modification is made in such a way that the distance value is set to a reduced, in particular minimal value, so that the safety during operation is increased overall.

As already mentioned above, the operating event to be detected is a predetermined operator movement which is recorded via the temporal course of the sensor measured values 8, 9.

As mentioned above, the operating event to be recorded is preferably a foot movement of the operator, whereby the sensor arrangement 5 has at least two, here exactly two, sensor elements 6, 7 designed as proximity sensors on which, as in 1 shown, are arranged in or on a rear panel part 11, here and preferably the rear bumper 11, of the motor vehicle. The two sensor elements 6, 7 are preferably capacitive proximity sensors with electrodes, each of which extends over a significant part of the width of the motor vehicle. This ensures that the detection of the operating event is possible across the significant part of the width of the motor vehicle.

The sensor arrangement 5 can be arranged at any location on the motor vehicle. It is conceivable, for example, that the sensor arrangement 5 is arranged in the side area of the motor vehicle, in particular on a fender, a side door or the like.

According to the illustration, 1 that the schematically shown detection area of the upper sensor element 6 also points backwards and the detection area of the lower sensor element 7 also points downwards. This sensor arrangement 5 leads accordingly to the 3 Sensor readings shown 8, 9.

As explained above, the predetermined operating event can vary considerably depending on the operator, so that operator-specific operating event models are proposed to be used in the operating event monitoring. In this case, and preferably in the context of the operating event monitoring, generally known methods of pattern recognition are used.

As a basis for pattern recognition, a series of characteristic features are assigned to the behavior of the sensor measured values 8, 9 of at least one sensor element 6, 7 that is characteristic of the operating event to be recorded. These characteristics are part of the operating event model and the occurrence of which is monitored as part of the operating event monitoring. As part of pattern recognition, these features are first extracted from the sensor measured values 8, 9 and then classified.

This corresponds to the basic concept of pattern recognition. In this process, pre-filtered measured values are first subjected to feature extraction. The features to be extracted must be selected in such a way that they are "separatively effective". This means that the features must be suitable for distinguishing the state "with operating event" from the state "without operating event" [ BISHOP, Christopher M.: Pattern recognition and machine learning. 5th ed. New York [ua] : Springer, 2007 (Information science and statistics). - ISBN 0-387-31073-8 ].

After feature reduction, a classification is carried out in which the formation of the extracted features is assigned to certain classes of patterns to be recorded.

At the 3 The forward and backward movement of the operator's foot shown generates pulse-like sensor events in the sensor measurement values 8, 9, which are referred to below as "sensor pulses". The width, height, edge steepness, curvature of parts of the sensor pulses, the time offset of two sensor pulses of a sensor element 6, 7 or the like can be defined as features to be extracted. As part of the classification, these features are then checked for limit values or limit ranges.

Other conceivable features here are the temporal offset of the sensor pulses of two sensor elements 6, 7 or a predetermined correlation, in particular the cross-correlation of the sensor measured values 8, 9 of two sensor elements 6, 7. These features can also be classified with the help of limit values and limit ranges in such a way that, if necessary, together with the previously mentioned features, a conclusion can be drawn about the occurrence or non-occurrence of an operating event.

The key point in the above pattern recognition is the fact that the series of features underlying the pattern recognition are assigned to the operator event model. When generating or parameterizing the operator event model, it is a matter of determining these features. The pattern recognition then runs in a known manner.

For all the embodiments explained above, it should be pointed out again that the control arrangement 4 can be arranged in a decentralized manner in terms of hardware. This can, for example, lead to at least part of the operator event model being stored in the sensor arrangement 5 and/or part of the operator event monitoring being carried out in the sensor arrangement 5.

According to a further teaching, which also has independent significance, the closure element arrangement 1 with closure element 2, drive arrangement 3, control arrangement 4 and sensor arrangement 5 is claimed as such. Reference may be made to all embodiments that are suitable for describing the closure element arrangement 1.

It has already been pointed out that both proposed teachings can be applied to all types of locking elements 2. In addition to the application to a tailgate 2 explained, the use on a sliding door can be highlighted as being particularly advantageous. Because sliding doors regularly only cover a small area of movement, the automatic functions described above can be implemented without a significant risk of collision. Against this background, the application of the proposed teachings to sliding doors is preferred.

Claims (13)

Method for controlling a closure element arrangement (1) of a motor vehicle, wherein the closure element arrangement (1) has a closure element (2), a drive arrangement (3) assigned to the closure element (2), a control arrangement (4) and a sensor arrangement (5) with at least one sensor element (6, 7), wherein, as part of an operating event monitoring, sensor events in the sensor measurement values (8, 9) of the sensor arrangement (5) are monitored by means of the control arrangement (4) to determine whether a predetermined operating event is present, and the drive arrangement (3) is controlled depending on the result of the operating event monitoring, wherein the operating event monitoring is based on at least one operating event model stored in the control arrangement (4) and wherein the operating event model is generated and/or parameterized in normal operation, characterized in that in the event that a sensor event was not recognized as an operating event as part of the operating event monitoring and that a control of the drive arrangement (3) was subsequently triggered manually, an operating event model is generated or parameterized from the sensor event by means of the control arrangement (4). such that the sensor event is recognized as an operating event during further operation. Procedure according to Claim 1 , characterized in that at least two potential operators of the motor vehicle are provided and that the control arrangement (4) determines which of the potential operators is active with regard to the operation of the motor vehicle. Procedure according to Claim 1 or 2 , characterized in that at least two potential operators of the motor vehicle are each assigned an individual radio key (10) and that the control arrangement (4) determines from the presence and/or operation of the radio key (10) which of the potential operators is active with regard to the operation of the motor vehicle. Method according to one of the preceding claims, characterized in that at least two operating event models are stored in the control arrangement (4), which are assigned to different operators, and that the operating event monitoring is based on the operating event model of the operator who is active in each case with regard to the operation of the motor vehicle. Method according to one of the preceding claims, characterized in that the control arrangement (4) can be brought into a learning mode in which the operator carries out an operating event, and that an operating event model is generated or parameterized from the resulting sensor measured values (8, 9) by means of the control arrangement (4) and then stored. Procedure according to Claim 5 , characterized in that the learning mode is carried out operator-specifically such that the operator event model is stored for a specific operator. Method according to one of the preceding claims, characterized in that, in the event that a sensor event was recognized as an operating event within the scope of the operating event monitoring, the operator active with regard to the operation of the motor vehicle is determined by means of the control arrangement (4) and the operating event model underlying the operating event is assigned to the operator active with regard to the operation of the motor vehicle. Method according to one of the preceding claims, characterized in that the operator event model is assigned to an operator. Method according to one of the preceding claims, characterized in that the operating event is defined as a predetermined operator movement which is detected via the temporal course of the sensor measured values (8, 9). Procedure according to Claim 9 , characterized in that at least one operating event is defined as a foot movement of the operator and that the sensor arrangement (5) has at least two sensor elements (6, 7) designed as proximity sensors, which are arranged in or on a rear panel part (11) of a motor vehicle. Procedure according to Claim 10 , characterized in that at least one operating event is defined as a back and forth movement of the operator's foot. Method according to one of the preceding claims, characterized in that the operating event model contains a series of features which are assigned to the course of the sensor measured values (8, 9) of at least one sensor element (6, 7) which is typical for the operating event and which are initially extracted from the sensor measured values (8, 9) in the manner of pattern recognition and then classified within the scope of the operating event monitoring. Closure element arrangement of a motor vehicle, wherein the closure element arrangement has a closure element (2), a drive arrangement (3) assigned to the closure element (2), a control arrangement (4) and a sensor arrangement (5) with at least one sensor element (6, 7), wherein, as part of an operating event monitoring, the control arrangement (4) monitors sensor events in the sensor measurement values (8, 9) of the sensor arrangement (5) to determine whether a predetermined operating event is present, and controls the drive arrangement (3) depending on the result of the operating event monitoring, wherein the operating event monitoring is based on at least one operating event model stored in the control arrangement (4) and wherein the operating event model can be generated and/or parameterized during normal operation, characterized in that in the event that a sensor event was not recognized as an operating event during the operating event monitoring and that a control of the drive arrangement (3) was subsequently triggered manually, an operating event model is generated or parameterized from the sensor event by means of the control arrangement (4) in such a way that the sensor event is recognized as an operating event during further operation. Operator event is detected.

Images