Tuning of gesture recognition algorithms for gesture-controlled opening of areas of a vehicle enclosed by moving elements

Abstract

According to a method for adjusting a gesture recognition algorithm for gesture-controlled opening of an area of a vehicle (1) closed by a movable element (3), a gesture recognition algorithm is provided in computer-readable form, the gesture recognition algorithm being adapted to recognize a predetermined gesture in response to a time-dependent sensor signal of an ultrasonic sensor (5) of the vehicle (1). A first time-dependent sensor signal is generated by the ultrasonic sensor (5) while a first person (8) performs a gestural movement in the field of view of the ultrasonic sensor (5), and a first value of a predetermined parameter of the gesture recognition algorithm is determined and stored in response to the first sensor signal.

Description

[Technical Field]

[0001] The present invention relates to a method for adjusting a gesture recognition algorithm for the gesture-controlled opening of an area of a vehicle closed by a moving element. The invention further relates to a method for gesture-controlled opening of an area of a vehicle closed by a moving element, and to a corresponding device for the vehicle, and to a computer program product. Summary of the Invention

[0002] Devices are known for automatically opening the trunk lid or tailgate of a vehicle when a user steps into, for example, a designated target area in the vicinity of the vehicle, and for this purpose, for example, a capacitive sensor is employed to detect the presence of a foot in the target area.

[0003] A disadvantage of this is that these devices react relatively inspecifically, which means that there is a risk of false positives, for example, if a user does not intend for the trunk lid or tailgate to be automatically opened and another object is brought into the target area. To prevent this, for example, the device can be configured so that the target area is appropriately small, thereby reducing the likelihood of an object being brought into the target area unintentionally. However, this is also a disadvantage because it means that the user must enter the relatively small target area if automatic opening of the trunk lid or tailgate is actually desired, which in turn reduces the reliability of recognizing the user's intention.

[0004] Furthermore, it is also known to use ultrasonic sensors for gesture recognition, since the distance of an object from the ultrasonic sensor can be evaluated in a time-dependent manner by the ultrasonic sensor, so that, for example, the time profile of a user's foot and, accordingly, the gestures that the user makes with his / her foot can also be recognized and characterized by the ultrasonic sensor.

[0005] However, when an ultrasonic sensor is used for gesture recognition, it is in principle possible that the corresponding user movement will not be recognized as a gesture, since the action may differ in detail, for example depending on the person or user.

[0006] Document DE 10 2016 013 935 A1 describes a method for automatically executing vehicle functions, in which a user is recognized via a vehicle key and then gestures and behavior patterns are recorded via one or more vehicle cameras. Vehicle settings can be personalized accordingly, for example, seat positions can be set in a correspondingly personalized manner.

[0007] The object of the present invention is to define the possibility of contactlessly opening areas of a vehicle closed by movable elements of the vehicle, which possibility allows the user's intention to open to be reliably recognized.

[0008] This object is achieved by the subject matter of the respective independent claims. Advantageous developments and preferred embodiments are the subject matter of the dependent claims.

[0009] The invention is based on the idea of adjusting a predetermined gesture recognition algorithm for ultrasound-based gesture recognition by determining and storing values of predetermined parameters of the gesture recognition algorithm depending on a sensor signal generated by an ultrasonic sensor of the vehicle while a person performs a gesture action.

[0010] According to one aspect of the invention, a method for adjusting a gesture recognition algorithm for gesture-controlled opening of an area enclosed by a movable element of a vehicle, particularly a motorized vehicle, is provided. The gesture recognition algorithm is provided in a computer-readable format. The gesture recognition algorithm is adapted to recognize a predetermined gesture in response to a time-dependent sensor signal of an ultrasonic sensor of the vehicle. A time-dependent first sensor signal is generated by the ultrasonic sensor while a first person performs a gestural motion within the field of view of the ultrasonic sensor. A first value of a predetermined parameter of the gesture recognition algorithm is determined, stored, and specifically adapted as a function, particularly by at least one computing unit.

[0011] The vehicle ultrasonic sensor and the optional further ultrasonic sensor may, for example, be part of an ultrasonic sensor system for the vehicle, which may also include at least one computing unit operable to activate the ultrasonic sensor and capable of determining and storing a first value of a predetermined parameter in response to the first sensor signal.

[0012] The ultrasonic sensor system can be operated, for example, in a training mode, and the first sensor signal can be generated in the training mode. In addition, the ultrasonic sensor system can be operated, for example, in a gesture recognition mode, particularly after the first value of the parameter has been determined and stored. The gesture recognition algorithm is particularly adapted to recognize a gesture in the gesture recognition mode in response to the sensor signal of the ultrasonic sensor.

[0013] The at least one computing unit may activate the ultrasonic sensor to emit ultrasonic waves, for example, to activate or deactivate the adjustment mode and / or the gesture recognition mode.

[0014] The ultrasonic wave can be emitted by the ultrasonic sensor, for example, in the form of one or more ultrasonic pulses. A portion of the emitted ultrasonic wave is reflected by the environment and detected by the ultrasonic sensor, prompting the ultrasonic sensor to generate a sensor signal or a first sensor signal, in particular as an electrical signal, a voltage signal or a current signal. The ultrasonic sensor may also be called an ultrasonic transducer.

[0015] The first sensor signal may for example correspond to the amplitude or intensity of the reflected and detected ultrasound waves in each case at a point in time, or to the corresponding envelope.

[0016] The first person is positioned within the field of view of the ultrasonic sensor and performs a gestural movement, so that a first sensor signal reflects or is generated in response to the gestural movement. The gestural movement can be interpreted as a movement of a body part of the first person, in particular a movement of a foot or a knee, which represents or is intended to represent or reproduce a predetermined gesture. During the performance of the gestural movement, the first person can be positioned, for example, in a predetermined gesture recognition area outside the vehicle, in particular in the immediate vicinity of the ultrasonic sensor, so that the field of view of the ultrasonic sensor covers the gesture recognition area.

[0017] After determining and saving the first values for the parameters, the adjustment mode can be deactivated, for example. Then, the gesture recognition mode can be activated, for example. The determination and storage, in particular adaptation, of the first values of the parameters of the gesture recognition algorithm can be understood as adjusting the gesture recognition algorithm. After the inventive method for adjusting the gesture recognition algorithm has been performed, the gesture recognition algorithm can be used to recognize a predetermined gesture depending on the first values of the parameters. In a similar manner, further parameters of the gesture recognition algorithm can also be determined and stored depending on the first sensor signal and can be used for further use of the gesture recognition algorithm in the gesture recognition mode.

[0018] Since the first value of the parameter is determined in a person-related manner in response to the gesture of the first person, characteristic movement patterns or characteristics of the gesture of the first person can be taken into account for subsequent gesture recognition in the gesture recognition mode. For example, a gesture recognition algorithm can, in principle, already detect a gesture in response to the sensor signal of an ultrasonic sensor even before the inventive adjustment method is performed. However, the inventive method takes into account the person-dependent characteristics of the gesture, and the gesture recognition mode is adapted accordingly. As a result, the reliability of gesture recognition in the gesture recognition mode can be increased. In other words, the possibility that the first person wants to perform a gesture that is not recognized by the gesture recognition algorithm due to the characteristic nature of the first person's movement can be reduced.

[0019] According to at least one embodiment of the method for adjusting a gesture recognition algorithm, characteristic properties of a gesture movement for a first person are determined in response to a first sensor signal, in particular by at least one computing unit, and a first value of a parameter is determined in response to the determined characteristic properties for the first person, in particular by at least one computing unit.

[0020] The characteristic property can for example be the initial distance of the body part of the first person that is moved during the gesture action, or the characteristic property can be derived therefrom. Thus, it can be advantageously taken into account that different people usually adopt different distances from the vehicle when performing the gesture action.

[0021] Similarly, the characteristic property may correspond to or be derived from the final spacing of the body parts after or upon completion of the gesture movement.

[0022] The gesture movement can correspond, for example, to a movement of a body part, in particular a foot, from a starting position towards the vehicle or the ultrasonic sensor and back or nearly back to the starting position, so that the distance of the body part from the ultrasonic sensor, in particular at the start of the gesture movement, is equal to an initial distance, then decreases, and then increases again until it reaches a final distance.

[0023] By considering the final interval, it is possible to take into account, for example, that different people usually move or take a step towards the ultrasonic sensor when performing a gesture, i.e., different people may not completely return their feet to the starting position and the initial interval, and therefore the final interval does not match the initial interval, which can be utilized for person-specific characterization of the first value for the parameter and for the corresponding person-specific identification.

[0024] The characteristic property may also correspond to or be derived from the distance between the final interval and the initial interval.

[0025] The characteristic property may correspond to or be derived from the speed of the movement of the body part during the gesture movement.

[0026] Thus, it can advantageously be taken into account that different people typically move body parts at different speeds to perform gesture actions.

[0027] This characteristic property can correspond to or be derived from the duration of the gesture, which can also be used to map and take into account person-specific characteristics of the gesture, such as different gestural speeds and person-specific body proportions such as leg length.

[0028] The characteristic property may also correspond to or be derived from the amplitude, in particular the maximum amplitude, of the signal pulses of the first sensor signal during the gesture movement, which may in particular be influenced by, for example, the body size, build or clothing of the first person, which are further typical variables influencing the first value of the feature or parameter of the gesture movement.

[0029] The characteristic properties are not limited to the examples described. Some or other of the characteristic properties described can be determined based on the first sensor signal, and the first value of the parameter can be determined accordingly. Two or more characteristic properties can be combined with each other or offset with each other to determine the first value for the parameter.

[0030] In general, the parameter may correspond directly to one of the characteristic properties or may be derived from one or more of the characteristic properties.

[0031] By taking characteristic features as a basis for specifying the first value for the parameter, the gesture recognition algorithm can be influenced in a targeted manner to increase the reliability of the gesture recognition. As a result, the first value of the parameter can be specified in particular so that the gesture recognition algorithm can take into account characteristic features that are particularly relevant during gesture recognition, and characteristic features that may be less relevant are given less weight or less during gesture recognition. In particular, this can prevent excessive requirements from being imposed on the gesture recognition algorithm and can prevent variations in gesture movements of the same person from affecting the reliability of the gesture recognition.

[0032] According to at least one embodiment, the generation of the first sensor signal is repeated while the first person repeatedly performs a gesture within the field of view of the ultrasonic sensor. The determination of a characteristic property of the gesture for the first person is also repeated based on the repeatedly generated first sensor signal. A first value of the parameter is determined in response to the repeatedly determined characteristic property, for example by averaging or other statistical processing.

[0033] This allows for variations in the characteristic characteristics of gestures made by the same person during their performance, thereby further increasing the reliability of gesture recognition.

[0034] According to at least one embodiment, the gesture recognition algorithm implements a state machine.

[0035] The state machine represents an algorithm that can establish one of two or more predefined states in response to the first sensor signal, for example of the first sensor signal or of the gesture movement or of the first person, and in particular can detect the presence of a gesture based on the time sequence of the states thus identified.

[0036] For example, the various states of the state machine can correspond to situations where the body part is stationary within a certain distance from the ultrasonic sensor and / or where the body part is moving in a certain direction. Thus, in a non-limiting example, a gesture may be detected when, for example, the body part is stationary in the gesture recognition area, followed by a state where the body part is moving toward the ultrasonic sensor, followed by a state where the body part is moving away from the ultrasonic sensor, followed by a state where the body part is stationary in the gesture recognition area.

[0037] For example, the predetermined parameter and the further parameter can be used to identify the individual state. For example, one or more characteristic properties can be used directly or can be treated as parameters for the state machine for recognizing the individual state. In this way, specifying first values for the parameters based on person-specific criteria can directly influence state recognition by the state machine, and thus gesture recognition.

[0038] According to at least one embodiment, a time-dependent second sensor signal is generated by the ultrasonic sensor while a second person, in particular different from the first person, performs a gesture within the field of view of the ultrasonic sensor. A characteristic property of the gesture for the second person is determined in response to the second sensor signal, in particular by the at least one computing unit. A second value of the parameter is determined and stored in response to the determined characteristic property for the second person, in particular by the at least one computing unit.

[0039] Thus, different values of parameters or different variations of gesture recognition algorithms assigned to different people can be saved and maintained. For example, corresponding user profiles can be created for various people, which save respective values for parameters or corresponding adapted gesture recognition algorithms. As a result, the reliability of gesture recognition can be further improved.

[0040] According to at least one embodiment, the gesture recognition algorithm includes a trained recurrent neural network RNN, for example a long short-term memory LSTM, and the parameters are weighting coefficients or bias parameters of the RNN.

[0041] In such an embodiment, the gesture recognition algorithm can directly recognize gestures without requiring an intermediate step of identifying characteristic features to be performed, for example, based on input data generated in response to sensor signals from an ultrasonic sensor, which has the advantage of not necessarily knowing person-specific differences in gesture movements and their effect on gesture detection.

[0042] In such an embodiment, a method for adjusting the gesture recognition algorithm can be understood to improve or refine the training of the RNN based on the first sensor signal.

[0043] For example, when the ultrasonic sensor system is in training mode, it can be assumed that the performed gesture movements actually correspond to the gesture, and the parameterization of the RNN can therefore be adapted as needed depending on the first sensor signal, for example to increase the reliability with which the gesture is detected.

[0044] According to at least one embodiment, a calibration mode of the ultrasonic sensor system is activated, for example, by at least one computing device, and a first sensor signal is generated in the calibration mode. The calibration mode can be initiated, for example, by a first person.

[0045] For example, at least one wireless signal can be transmitted by an electronic device external to the vehicle to the at least one computing device, and the adjustment mode can be activated in response to the transmitted at least one wireless signal.

[0046] An electronic device that is external to the vehicle is therefore not particularly part of the vehicle but may for example correspond to a mobile electronic terminal, a smartphone, a smart watch or other so-called wearable device, or a vehicle key or key chain for the vehicle, etc. Thus, an adjustment mode can be initiated, in particular if the first person notices that the reliability of the gesture detection needs to be improved.

[0047] According to at least one embodiment, an identification of a first person is transmitted to at least one computing device using at least one wireless signal, and a first value of a parameter is stored in a user profile of the first person according to the identification, e.g., in a corresponding embodiment, the same for a second person.

[0048] In various embodiments, the adjustment mode may also be activated in response to user input recorded by a user input device of the vehicle.

[0049] According to a further aspect of the invention, a method for gesture-controlled opening of an area of a vehicle enclosed by a movable element is provided. A method according to the invention for adjusting a gesture recognition algorithm is implemented here. After the method for adjusting the gesture recognition algorithm is implemented, a time-dependent further first sensor signal is generated by the ultrasonic sensor while a first person again performs a gesture motion within the field of view of the ultrasonic sensor, e.g., during gesture recognition mode. By applying the gesture recognition algorithm to the input data in response to the further sensor signal, a predetermined gesture is detected, particularly by at least one computing unit, in response to the stored first value. In response to the detection of the gesture, the movable element is automatically moved to open the enclosed area.

[0050] In response to detecting the gesture, it can be understood that automatic movement of the movable element occurs only if the gesture is detected. In other words, input data is generated in response to the further first sensor signal, and a gesture recognition algorithm is applied to the input data in response to the stored first value of the parameter to detect the gesture.

[0051] The gesture recognition mode can be activated, for example, when a first person is located in the gesture recognition area, but further conditions for activating the gesture recognition mode can also be envisaged.

[0052] The further first sensor signal reflects the distance of the first person or a body part of the first person from the ultrasonic sensor, so that the performance of the gesture action can be recognized as a gesture and detected accordingly.

[0053] Depending on the embodiment of the method and the vehicle, the movable element can be, for example, a door, in particular a side door, a driver's door, a passenger door, a rear door, a back door, a sliding door, a tailgate or a trunk lid. It is preferably a tailgate or a trunk lid. In this case, the closed area of the vehicle is, for example, the trunk or a loading area of the vehicle that can be opened or closed. The open loading area can be open at the top, in particular, it does not exclude that the open loading area can be closed in the rear direction by a tailgate, as is the case, for example, in the case of small vans or pick-up trucks with an open loading surface.

[0054] For application cases or situations that may arise in the method and that are not explicitly described here, it is possible to envisage that, according to the invention, an error message and / or a request for input of user feedback is output and / or default settings and / or pre-specified initial states are set.

[0055] According to a further aspect of the invention, a device for a vehicle for gesture-controlled opening of an area of the vehicle enclosed by a movable element is provided, the device comprising an ultrasonic sensor system with an ultrasonic sensor configured to generate a time-dependent first sensor signal by the ultrasonic sensor while a first person performs a gesture motion within a field of view of the ultrasonic sensor, and at least one computing unit configured to determine and store a first value of a predetermined parameter of a gesture recognition algorithm in response to the first sensor signal.

[0056] A computing unit may be understood to be, in particular, a data processing device including processing circuits. The computing unit is therefore capable of processing data, in particular for performing computations. Optionally, it may also include operations for performing indexed accesses to data structures, for example look-up tables (LUTs).

[0057] The computing unit may in particular comprise one or more computers, one or more microcontrollers, and / or one or more integrated circuits, such as one or more application specific integrated circuits (ASICs), one or more field programmable gate arrays (FPGAs), and / or one or more systems on a chip (SoCs). The computing unit may also comprise one or more processors, such as one or more microprocessors, one or more central processing units (CPUs), one or more graphic processing units (GPUs), and / or one or more signal processors, in particular one or more digital signal processors (DSPs). The computing unit may also comprise a group of physical or virtual computers or other types of the above-mentioned units.

[0058] In various exemplary embodiments, the computing unit includes one or more hardware and / or software interfaces and / or one or more storage devices.

[0059] The memory device can be configured as a volatile data memory, for example as a dynamic random access memory (DRAM) or a static random access memory (SRAM), or as a non-volatile data memory, for example as a read only memory (ROM), as a programmable read only memory (PROM), as an erasable programmable read only memory (EPROM), as an electrically erasable programmable read only memory (EEPROM), as a flash memory or flash EEPROM, as a ferroelectric random access memory (FRAM), as a magnetoresistive random access memory (MRAM), or as a phase change random access memory (PCRAM).

[0060] According to at least one embodiment, the device includes a storage medium storing a gesture recognition algorithm in computer-readable form. The ultrasonic sensor system is designed to generate a time-dependent further first sensor signal by the ultrasonic sensor after determining and storing a first value of the parameter while the first person again performs a gesture motion within the field of view of the ultrasonic sensor. The at least one computing unit is designed to apply the gesture recognition algorithm to input data, the input data being a function of the further first sensor signal, in response to the stored first value of the parameter to detect a predetermined gesture. The at least one computing unit is designed to generate at least one control signal to automatically move a movable element in response to detecting the gesture.

[0061] According to at least one embodiment, the device comprises at least one actuator designed to automatically move a movable element in response to a control signal to open the enclosed area.

[0062] The at least one computing unit can send the control signal directly to the at least one actuator or to a control unit for the at least one actuator, which generates a further control signal and sends it to the at least one actuator to prompt it to move the movable element in response to the further control signal.

[0063] When this disclosure refers to a component of a device according to the invention, and in particular to at least one computing unit of the device, as being designed, formed, configured, etc. to perform or realize a certain function, achieve a certain effect, or fulfill a certain purpose, this can be understood as the component being specifically and actually capable of performing or realizing that function, achieving that effect, or fulfilling that purpose by appropriate adaptation, programming, physical configuration, etc., beyond the component's basic or theoretical usability or suitability for that function, effect, or purpose.

[0064] Further embodiments of the apparatus according to the invention can be derived directly from various configurations of the method according to the invention, and vice versa. In particular, individual features and corresponding explanations and advantages relating to various embodiments of the method according to the invention can be transferred in an analogous manner to corresponding embodiments of the apparatus according to the invention. In particular, the apparatus according to the invention is configured or programmed to carry out the method according to the invention. In particular, the apparatus according to the invention carries out the method according to the invention.

[0065] According to a further aspect of the invention, a computer program is provided which comprises instructions which, when executed by an apparatus according to the invention, in particular by at least one computing unit, cause the apparatus to perform a method according to the invention for adjusting a gesture recognition algorithm and / or a method according to the invention for gesture-controlled opening of an enclosed area of a vehicle.

[0066] According to a further aspect of the invention, a computer readable storage medium storing a computer program according to the invention is provided.

[0067] The computer program and the computer-readable storage medium may be construed as respective computer program products containing instructions.

[0068] Further features of the invention can be found in the claims, the drawings, and the description of the drawings. Features and combinations of features mentioned above in the specification, as well as features and combinations of features mentioned below in the description of the drawings and / or shown in the drawings, may be included in the invention not only in the combination specified in each case, but also in other combinations. In particular, embodiments and combinations of features that do not comprise all the features of the originally worded claims may also be included in the invention. Furthermore, embodiments and combinations of features that go beyond or differ from the combinations of features set out in the later references to the claims may also be included in the invention.

[0069] The invention will now be described in more detail on the basis of specific exemplary embodiments with reference to the associated schematic drawings, in which identical or functionally identical elements may be given the same reference signs, and descriptions of identical or functionally identical elements will not necessarily be repeated with respect to different figures. [Brief explanation of the drawings]

[0070] In the drawings: [Figure 1] FIG. 1 shows a schematic view of a vehicle equipped with an exemplary embodiment of a device for a vehicle for the automatic opening of an area of the vehicle closed by a mobile element according to the invention. [Figure 2] FIG. 2 is a schematic flow chart of an exemplary embodiment of a method for the automatic opening of an area of a vehicle closed by a mobile element according to the invention. DETAILED DESCRIPTION OF THE INVENTION

[0071] FIG. 1 shows diagrammatically a vehicle 1 equipped with an exemplary embodiment of a device 2 according to the invention for gesture-controlled opening of an area of the vehicle 1 closed by a mobile element 3 .

[0072] The movable element 3 is depicted here by way of example as a trunk lid of the vehicle 1, and therefore the area closed by it is the trunk of the vehicle 1. In other embodiments, the movable element may instead be a tailgate or a door of the vehicle 1.

[0073] The vehicle, e.g., device 2, comprises, in particular, one or more actuators 4a, 4b, two actuators 4a, 4b in the non-limiting example of FIG. 1 , and a computing unit 6. The computing unit 6 can be considered representative of one or more control units, computing units, and / or evaluation units, which are described in combined functional terms for ease of explanation. The computing unit 6 can, in particular, activate the actuators 4a, 4b, which, for example, automatically initiate the movement of the movable element 3 and, optionally, unlock the movable element 3. As a result, the area of the vehicle thus closed by the movable element 3 can be opened automatically, i.e., without the user having to manually open the movable element 3. Systems in which this can be unlocked by the vehicle key of the vehicle 1 are known. This may also be possible in the present device, but is not strictly necessary. Rather, device 2 can implement the method according to the invention for gesture-controlled opening of a closed area.

[0074] For this purpose, the device 2 has an ultrasonic sensor 5, which has a field of view capable of detecting objects, such as a person 8. In particular, the ultrasonic sensor 5 is capable of transmitting ultrasonic pulses, detecting portions of the ultrasonic pulses reflected by objects within the field of view of the ultrasonic sensor 5, and generating corresponding time-dependent sensor signals that reproduce the intensity of the detected portions of the ultrasonic pulses. The sensor signal typically includes signal pulses, also called peaks or echoes, each of which is based on an emitted ultrasonic pulse. The temporal positions of the signal pulses allow the distance of the reflecting object from the ultrasonic sensor 5 to be determined by a computing unit 6.

[0075] The ultrasonic sensor 5 and the computing unit 6 or parts of the computing unit 6, as well as possibly further ultrasonic sensors (not shown), can be considered as part of an ultrasonic sensor system of the device 2. The ultrasonic sensor system, in particular the ultrasonic sensor 5, is not necessarily usable only in the sense of the invention, but can additionally optionally also be used for other functions, such as for parking assistance for the driver of the vehicle 1.

[0076] The method according to the invention for gesture-controlled opening of an area of a vehicle 1 closed by a movable element 3 provides for the implementation of the method according to the invention for adjusting a gesture recognition algorithm, which is stored in particular in a computer-readable form on a storage medium of the computing unit 6.

[0077] In the calibration mode, which can be initiated by the person 8, for example via a user input interface of the vehicle 1 or an electronic device external to the vehicle, such as a smartphone, the person 8 can make a gestural movement in the field of view of the ultrasonic sensor 5, for example in a predetermined gesture recognition area 7 in the vicinity of the ultrasonic sensor 5, such as moving their foot from a standing position towards the ultrasonic sensor 5 and back again. While the person 8 is making the gestural movement, the ultrasonic sensor 5 generates a corresponding time-dependent sensor signal. As part of its function, the computing unit 6 determines and stores values of predetermined parameters of the gesture recognition algorithm, particularly specifically for the person 8.

[0078] For this purpose, the computing unit 6 can determine characteristic properties of the gesture movement of the person 8, for example in response to the sensor signals, and determine the value of the parameter in response to the characteristic properties. Optionally, the person 8 can also perform the gesture movement repeatedly in order to compensate for variations in the characteristic properties.

[0079] After completion of the method for adjusting the gesture recognition algorithm, the same person 8 can appear in the gesture recognition area 7 at a later time in gesture recognition mode and perform a gesture action again. The adjusted gesture recognition algorithm can generate input data for the gesture recognition algorithm in response to the correspondingly generated further sensor signal of the ultrasonic sensor, and can apply the gesture recognition algorithm to the input data, in particular with or in response to the stored values of the parameters, so that the gesture recognition algorithm can detect the predetermined gesture.

[0080] If a gesture is detected, the computing unit 6 is able to generate at least one control signal for an automatic movement of the movable element 3, which is sent to the actuators 4a, 4b, thereby initiating the opening of the closed area.

[0081] For the gesture recognition itself, the computing unit 6 can use methods known per se. For example, the gesture recognition algorithm can implement a state machine. Thus, by applying the gesture recognition algorithm to the input data, the computing unit 6 can recognize when the gesture movements made by the person 8 correspond to a predetermined sequence of two or more states, and based on that, can detect the gesture.

[0082] However, other algorithms for gesture recognition are also contemplated and may be used, for example, the gesture recognition algorithm may include a trained recurrent neural network (RNN).

[0083] 2 shows a schematic flow chart of an exemplary embodiment of the method according to the invention for gesture-controlled opening of an area of the vehicle 1 closed by a movable element 3. According to steps S1, S2 and S3, an embodiment of the method according to the invention for adjusting the gesture recognition algorithm is carried out, optionally repeated for several people.

[0084] In particular, in step S1, a gesture recognition algorithm is provided in computer-readable form. In step S2, time-dependent sensor signals are generated while a person 8 is positioned within the field of view of the ultrasonic sensor 5 and performs a gesture. In step S3, parameter values are determined and stored in response to the sensor signals, e.g., by identifying characteristic characteristics of the gesture for the person 8. After steps S1-S3 have been performed, the training mode is terminated and, e.g., gesture recognition mode is activated at a later time. Thereafter, in step S4, a further sensor signal is generated by the ultrasonic sensor 5 while the person 8 again performs a gesture within the field of view of the ultrasonic sensor 5. In step S5, the gesture recognition algorithm is applied to input data that is a function of the further sensor signal in response to the stored parameter values, and a predetermined gesture is detected. In step S6, in response to the detection of the gesture, the movable element 3 is automatically moved to open the enclosed area.

[0085] As a result, particularly robust gesture recognition can be achieved, allowing for the recognition of gesture movements of various types and expressions. The sensor signals can, for example, be fed directly into an RNN, which can make a binary decision as to whether the gesture movement corresponds to a gesture or not.

[0086] The advantages of RNNs include the fact that they characterize an internal state, which acts like a memory for previous measurements so that changes between measurements can be recognized. This also eliminates the need for an external memory for the measurements, resulting in a saving in storage capacity. The RNN used can be, for example, an LSTM.

[0087] The RNN can be trained using the captured data to learn internal parameters and to recognize gestures, if any. Because sensor locations and other properties may vary depending on the vehicle model, it may be advantageous to tune the pre-trained RNN to the specific model of vehicle 1. This can be achieved by obtaining a dataset specific to this vehicle model and further training the RNN accordingly.

Claims

1. 1. A method for adjusting a gesture recognition algorithm for gesture-controlled opening of an area of a vehicle (1) closed by a movable element (3), comprising: - the gesture recognition algorithm is provided in a computer readable format adapted to recognize predetermined gestures in response to time-dependent sensor signals of an ultrasonic sensor (5) of the vehicle (1); - a time-dependent first sensor signal is generated by said ultrasonic sensor (5) while a first person (8) performs a gesture movement in the field of view of said ultrasonic sensor (5); a first value of a predetermined parameter of the gesture recognition algorithm is determined and stored in response to the first sensor signal; A method characterized by:

2. - a characteristic property of the gesture movement of the first person (8) is determined in response to the first sensor signal; - said first value of said parameter is determined as a function of said characteristic property determined for said first person (8); 2. The method of claim 1 .

3. The characteristic properties are: - related to the initial distance from the ultrasonic sensor (5) of the body part moved during the gesture; and / or - related to the final spacing of said body parts after or upon completion of said gesture movement; and / or - related to the speed of movement of said body part during said gesture; and / or - related to the duration of said gesture; and / or - related to the maximum amplitude of a signal pulse of the first sensor signal during the gesture movement, 3. The method of claim 2.

4. - the generation of the first sensor signal is repeated while the first person (8) repeatedly performs the gesture movement in the field of view of the ultrasonic sensor (5); - repeatedly identifying the characteristic properties of the gesture movement of the first person (8) based on the repeatedly generated first sensor signals; the first value of the parameter is determined as a function of the characteristic property that is repeatedly determined; 4. The method according to claim 2 or 3.

5. the gesture recognition algorithm implements a state machine; 5. The method according to claim 2,

6. - a time-dependent second sensor signal is generated by said ultrasonic sensor (5) while a second person is performing said gesture movement in the field of view of said ultrasonic sensor; - the characteristic properties of the gesture movement relating to the second person are determined in response to the second sensor signal; a second value of said parameter is determined and stored as a function of said characteristic property determined for said second person; 6. The method according to any one of claims 2 to 5.

7. the gesture recognition algorithm includes a recurrent neural network, and the parameters are weighting coefficients or bias parameters of the recurrent neural network; 2. The method of claim 1 .

8. A training mode of an ultrasonic sensor system (5, 6) including the ultrasonic sensor (5) is activated, and the first sensor signal is generated in the training mode.

8. The method according to any one of claims 1 to 7.

9. at least one radio signal is transmitted by an electronic device external to the vehicle to at least one computing unit (6) of the ultrasonic sensor system (5, 6); - said adjustment mode is activated in response to said at least one transmitted radio signal; 9. The method of claim 8.

10. - the identification information of said first person (8) is transmitted to said at least one computing unit (6) using said at least one radio signal; - said first value of said parameter is stored in a user profile of said first person (8) depending on said identification information; 10. The method according to claim 8 or 9.

11. the adjustment mode is activated in response to a user input recorded by a user input device of the vehicle (1); 9. The method of claim 8.

12. A method for gesture-controlled opening of an area of a vehicle (1) closed by a movable element (3), comprising: - a method for adjusting a gesture recognition algorithm is performed according to any one of claims 1 to 11; - after the method for adjusting the gesture recognition algorithm has been performed, a further time-dependent first sensor signal is generated by the ultrasonic sensor (5) while the first person (8) again performs the gesture movement in the field of view of the ultrasonic sensor (5); - detecting a predetermined gesture as a function of the stored first values of the parameters by applying the gesture recognition algorithm to input data that is a function of the further first sensor signal; - the movable element (3) is automatically moved upon detection of the gesture in order to open the enclosed area; A method characterized by:

13. A device (2) on a vehicle (1) for gesture-controlled opening of an area of the vehicle (1) closed by a movable element (3), comprising: - the device (2) comprises a storage medium storing a gesture recognition algorithm in computer readable form adapted to recognize a predetermined gesture in response to a time-dependent sensor signal of an ultrasonic sensor (5) of the vehicle (1); - the device (2) has an ultrasonic sensor system (5, 6) with an ultrasonic sensor (5), the ultrasonic sensor system (5, 6) being arranged to generate a time-dependent first sensor signal by the ultrasonic sensor (5) while a first person (8) performs a gesture movement in the field of view of the ultrasonic sensor (5); the ultrasonic sensor system (5, 6) comprises at least one computing unit (6) arranged to determine and store a first value of a predetermined parameter of a gesture recognition algorithm in response to the first sensor signal; An apparatus characterized in that

14. the ultrasonic sensor system (5, 6) is arranged to generate, after determining and storing the first value of the parameter, a time-dependent further first sensor signal by the ultrasonic sensor (5) while the first person (8) again performs the gesture movement in the field of view of the ultrasonic sensor (5); - said at least one computing unit (6) is arranged to apply said gesture recognition algorithm to input data which is a function of said further first sensor signals in dependence on said stored first values of said parameters in order to detect a predetermined gesture; - said at least one computing unit (6) is arranged to generate at least one control signal in response to detection of said gesture so as to automatically move said movable element (3); 14. Device (2) according to claim 13.

15. A computer program product comprising instructions which, when executed by an apparatus (2) according to claim 13, cause the apparatus (2) to perform the method according to any one of claims 1 to 11, and / or instructions which, when executed by an apparatus (2) according to claim 14, cause the apparatus (2) to perform the method according to claim 12.

Images