Method for improving object recognition, computer program, device and vehicle

Abstract

Exemplary embodiments of the present invention provide a method for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor. The method comprises obtaining vehicle data indicative of the vehicle's interior and obtaining ultra-wideband sensor data indicative of a specification of the ultra-wideband sensor. Furthermore, the method comprises determining a correction parameter for correcting the measurement signal of the ultra-wideband sensor for object detection. The method also includes storing the correction parameter to perform the correction of the measurement signal of the ultra-wideband sensor for object detection.

Description

[0001] Exemplary embodiments of the present invention relate to methods for improving object recognition, a computer program, a device and a vehicle, in particular but not exclusively to a concept for correcting a measurement signal of an ultra-wideband sensor.

[0002] Several aspects must be considered when specifying an ultra-wideband sensor with radar and radio functionality. On the one hand, the ultra-wideband sensor must be designed to detect an object, such as a living being, inside a vehicle using its radar functionality. On the other hand, the ultra-wideband sensor must be designed to communicate with the vehicle's digital key using its radio functionality. For cost reasons, both functionalities should ideally be integrated into a single ultra-wideband sensor.

[0003] DE 10 2023 101 419 A1 discloses a method for providing data for monitoring a vehicle interior. The method comprises receiving (S110) first data (111) that specifies raw data from a radar sensor (200) relating to the vehicle interior and exhibiting a first dimensionality. Furthermore, the method comprises performing (S120) a dimensionality reduction process on the first data (111) to generate second data (112) that exhibit a lower second dimensionality compared to the first data (111). The method also comprises providing (S130) the second data (112) to a vehicle bus (300).

[0004] DE 10 2022 208 095 A1 discloses a method 100 for a communication device for determining a sensor functionality of a vehicle. The method 100 comprises receiving 110 from a communication device an error signal indicative of a fault of a sensor of the vehicle and determining 120 a malfunction situation for the sensor of the vehicle based on the error signal.

[0005] There is therefore a need to provide a method for improving object detection based on a measurement signal from an ultra-wideband sensor. The method, the device, the computer program, and the vehicle according to the independent claims address this need.

[0006] Examples of implementations are based on the core idea that a method for object detection in the interior of a vehicle can be improved by a correction parameter for correcting the measurement signal of the ultra-wideband sensor. This allows, for example, a correction of the measurement signal, which can then be improved in the evaluation of the measurement signal, for example, an evaluation using a machine learning algorithm.

[0007] Exemplary embodiments relate to a method for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor. The method includes obtaining vehicle data indicative of the vehicle's interior and obtaining ultra-wideband sensor data indicative of a specification of the ultra-wideband sensor. Furthermore, the method includes determining a correction parameter to correct the measurement signal of the ultra-wideband sensor for object detection. The method also includes storing the correction parameter to perform the correction of the measurement signal of the ultra-wideband sensor for object detection. The correction parameter can simplify the evaluation of the measurement signal of the ultra-wideband sensor. Using the vehicle data and the ultra-wideband sensor data, the correction parameter can be adapted to a specific application. That is to say,The correction parameter can be used to correct the measurement signal of a specific ultra-wideband sensor in a particular vehicle. Accordingly, areas of the measurement signal that are irrelevant or disruptive for evaluation can be discarded. This improves the evaluation of the ultra-wideband sensor's measurement signal. The improved evaluation, in turn, enhances object detection.

[0008] The procedure further includes receiving the measurement signal from the ultra-wideband sensor, reading the correction parameter, and correcting the measurement signal by applying the correction parameter. Applying the correction parameter allows the measurement signal from the ultra-wideband sensor to be processed appropriately for evaluation. For example, the measurement signal, or a portion thereof, can be reduced.

[0009] In one embodiment, the correction can involve truncating measurement data from the measurement signal. The truncated measurement data may lie outside a propagation time of the measurement signal determined by the correction parameter. This allows, in particular, the truncation of events contained in the measurement signal that originate from outside the vehicle. This can reduce or prevent the erroneous detection of an object located outside the vehicle. Truncating the measurement data can therefore improve the focus of the measurement signal evaluation on measured values ​​from inside the vehicle.

[0010] In one embodiment, the method can further include obtaining additional ultra-wideband sensor data indicative of a specification for another ultra-wideband sensor. Furthermore, the method can include determining an additional correction parameter to correct another measurement signal from the additional ultra-wideband sensor for object detection and storing the additional correction parameter to perform the correction of the additional measurement signal from the additional ultra-wideband sensor for object detection. The additional correction parameter allows for the correction of another measurement signal. This allows, in particular, redundancy to be generated by using different measurement signals. The use of multiple correction parameters can enable specific adaptation of the measurement signals from different ultra-wideband sensors. This can improve object detection based on multiple ultra-wideband sensors.

[0011] In one embodiment, the method can further include receiving an additional measurement signal from the additional ultra-wideband sensor and reading out the additional correction parameter. The method can further include correcting the additional measurement signal by applying the additional correction parameter and performing object detection based on the corrected measurement signal and the corrected additional measurement signal. Using multiple measurement signals can improve object detection. In particular, irregularities in a measurement signal that could lead to faulty object detection can be identified or at least discarded.

[0012] In one embodiment, the object detection process can include comparing the corrected measurement signal with the corrected secondary measurement signal. Furthermore, if an anomaly between the corrected measurement signal and the corrected secondary measurement signal is identified during the comparison, the corrected measurement signal and / or the corrected secondary measurement signal can be discarded. An anomaly could, for example, be a deviation of the measurement signal from the secondary measurement signal that could lead to an inconsistency in object detection. Accordingly, object detection can be improved by discarding the corrected measurement signal and / or the corrected secondary measurement signal.

[0013] In one embodiment, the method can further include identifying a reproducible anomaly within a portion of the measurement signal and discarding that portion of the signal exhibiting the anomaly. A reproducible anomaly could, for example, be a periodic signal within the measurement signal whose periodicity does not correspond to that of an object inside the vehicle. By discarding the portion of the measurement signal containing the reproducible anomaly, object detection can be improved.

[0014] Exemplary embodiments also include a computer program for carrying out one of the methods described herein, if the computer program runs on a computer, a processor, or a programmable hardware component.

[0015] Another embodiment is a device for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor. The device comprises a communication interface and a data processing circuit configured to perform at least one of the methods described herein. Further embodiments provide a vehicle equipped with a device as described herein.

[0016] Examples of implementation are explained in more detail below with reference to the accompanying figures. These show: Fig. Figure 1 shows a schematic representation of a method 100 for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor; and Fig. Figure 2 shows a block diagram of an embodiment of a device for controlling a diagnostic session of a vehicle.

[0017] Several embodiments are now described in more detail with reference to the accompanying drawings, in which some of these embodiments are illustrated. For the sake of clarity, the thickness dimensions of lines, layers, and / or regions may be exaggerated in the figures.

[0018] Fig. Figure 1 shows a schematic representation of a method 100 for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor. The method 100 can be carried out by a device as described below, for example, by a central control unit of a vehicle. The method 100 comprises obtaining 110 vehicle data indicative of the interior of the vehicle. Obtaining 110 the vehicle data can include determining the vehicle data. For example, the device can be configured to read a configuration of the interior of the vehicle, such as a cover, from a memory location of the ultra-wideband sensor. The configuration of the interior can influence the propagation of a measurement signal emitted by the ultra-wideband sensor. In particular, the propagation time of the emitted measurement signal can depend on the interior of the vehicle. That is to say,The vehicle data can be indicative of an expected detectable measurement signal. Optionally or alternatively, receiving the vehicle data can include receiving the vehicle data. For example, a device can receive the vehicle data from a server.

[0019] Method 100 further comprises obtaining 120 ultra-wideband sensor data indicative of a specification for the ultra-wideband sensor. A specification can describe the final technical and functional characteristics of the ultra-wideband sensor. For example, the specification can serve as a reference for a user who wants to integrate the sensor into a system or application.

[0020] Receiving the ultra-wideband sensor data can include determining the ultra-wideband sensor data. For example, the device can be configured to read the ultra-wideband sensor's specifications from a memory. Optionally or alternatively, receiving the ultra-wideband sensor data can include receiving the ultra-wideband sensor data. For example, a device can receive the ultra-wideband sensor data from the ultra-wideband sensor. Based on the vehicle data and the ultra-wideband sensor data, an evaluation range for the measurement signal can be determined. For example, the entire measurement signal or only a portion of it, depending on the signal's propagation time, may be relevant for object detection. That is, there may be a need to adapt the measurement signal to the vehicle, i.e., its geometry (e.g., interior dimensions, window positions) and the ultra-wideband sensor.

[0021] Accordingly, the procedure further comprises determining a correction parameter for correcting the measurement signal of the ultra-wideband sensor for object detection. The correction parameter can be used to adjust the measurement signal for evaluation. That is, object detection based on the measurement signal of the ultra-wideband sensor can be improved by means of the correction parameter. Determining the correction parameter can contribute to improving object detection by providing the correction parameter for correction.

[0022] The procedure further includes storing the correction parameter (140) to correct the measurement signal of the ultra-wideband sensor for object detection. That is, the correction parameter can be made available for later correction of a measurement signal by storing it (140). Accordingly, object detection can be improved by determining (130) and storing the correction parameter (140).

[0023] Method 100 further comprises receiving the measurement signal from the ultra-wideband sensor. Receiving the measurement signal can include determining the measurement signal. For example, the device can monitor the ultra-wideband sensor and determine the measurement signal. Optionally or alternatively, receiving the measurement signal can include receiving it. For example, the device can receive the measurement signal. Method 100 further comprises reading the correction parameter and correcting the measurement signal by applying the correction parameter. By applying the correction parameter, the measurement signal from the ultra-wideband sensor can be appropriately adjusted for evaluation. For example, the measurement signal or a portion of the measurement signal can be reduced.

[0024] For example, the measurement signal can be processed using geofencing. Geofencing a measurement signal refers to the use of virtual boundaries (so-called geofences) to control the acquisition, monitoring, or analysis of measurement signals within a specific time range. This means that specific time intervals or time windows within the signal propagation time can be defined, which can be referred to as geofences. The measurement signal of the ultra-wideband sensor can be divided into individual segments or taps. A geofence can then be used to select and / or exclude specific segments or taps. In other words, the correction parameter can be indicative of the geofence. Using the correction parameter or the geofence, for example, a tap can be clipped or a tap can be selected. In this case, the measurement signal can include or be a channel impulse response.For example, a distance measurement within a channel impulse response can be limited. This allows the measurement signal to be adapted to a specific vehicle. For instance, taps representing the exterior of a vehicle can be omitted from the measurement signal, or taps representing the interior can be selected. This reduces the probability of erroneous object detection based on an object being identified as being outside the vehicle.

[0025] In one embodiment, the correction can involve truncating measurement data from the measurement signal. The truncated measurement data can lie outside a propagation time of the measurement signal defined by the correction parameter. That is, the correction parameter can define a geofence that is indicative of a maximum propagation time used by the measurement signal. Accordingly, segments or taps of the measurement signal that are not intended for object detection can be truncated.

[0026] In one embodiment, the method 100 can further comprise obtaining additional ultra-wideband sensor data indicative of a specification for another ultra-wideband sensor. Furthermore, the method 100 can comprise determining another correction parameter to correct another measurement signal from the additional ultra-wideband sensor for object detection and storing the additional correction parameter to perform a correction of the additional measurement signal from the additional ultra-wideband sensor for object detection. By using multiple measurement signals, redundancy can be created, which makes it possible to improve object detection. In particular, object detection using different ultra-wideband sensors can yield different results. Therefore, combining multiple measurement signals can increase reliability.

[0027] In one embodiment, the method 100 can further comprise receiving a further measurement signal from the further ultra-wideband sensor and reading out the further correction parameter. The method can further comprise correcting the further measurement signal by applying the further correction parameter and performing object detection based on the corrected measurement signal and the corrected further measurement signal. For example, an object can only be considered uniquely detected if an object inside the vehicle has been detected based on both the corrected measurement signal and the corrected further measurement signal.

[0028] Optionally or alternatively, the additional measurement signal can be used to determine whether the object is located inside the vehicle or not. For example, an object may have been detected by evaluating the corrected measurement signal. However, it may not have been possible to determine precisely whether the object is inside the vehicle or not. Accordingly, the additional corrected measurement signal can be used to verify whether the object is inside the vehicle. The ultra-wideband sensor and the additional ultra-wideband sensor can be positioned at different locations within the vehicle, allowing them to cover different areas. By using the corrected additional measurement signal, it can then be determined whether the object is indeed inside the vehicle. For example, the corrected additional measurement signal may be less sensitive to an exterior environment, or...The corrected measurement signal is less sensitive to the outside environment. Therefore, if the corrected measurement signal does not allow for object detection, it can be concluded that the object is not inside the vehicle.

[0029] In one embodiment, performing object detection can include comparing the corrected measurement signal with the corrected secondary measurement signal. Furthermore, if an anomaly between the corrected measurement signal and the corrected secondary measurement signal is identified during the comparison, the method can include discarding the corrected measurement signal and / or the corrected secondary measurement signal. For example, individual segments or taps of the corrected measurement signal and the corrected secondary measurement signal may be related to each other. That is, correlations may appear in certain taps of the two measurement signals. Therefore, if an anomaly exists in a tap of one measurement signal that is not present in the corresponding tap of the other measurement signal, the tap can be considered irrelevant. Thus, object detection can be improved by comparing multiple measurement signals.

[0030] An anomaly in the measurement signal of an ultra-wideband sensor can be caused by strong signal reflections, interference, distorted signals, loss of signal strength, multipath propagation, and / or temporal instabilities. In other words, an anomaly in the measurement signal of an ultra-wideband sensor can be an unusual or unexpected deviation in the signal, which may indicate reflections, interference, signal loss, multipath propagation, and / or temporal instabilities.

[0031] In one embodiment, the method 100 can further comprise identifying a reproducible anomaly within a portion of the measurement signal and discarding the portion of the measurement signal that exhibits the reproducible anomaly. A reproducible anomaly can, for example, be a periodic signal within the measurement signal whose periodicity does not correspond to an object inside the vehicle. For instance, a person may be moving around the outside of the vehicle. As soon as the person walks past a window of the vehicle, the measurement signal may contain a stronger signal from that person than when the person is shielded by the vehicle's body. That is, the reproducible anomaly can, for example, be an object that is only detectable in the signal at certain intervals, such as when the person walks past a window.This reproducible anomaly can be used to discard the measurement signal, or parts of it, for evaluation. This can improve object detection.

[0032] The measurement signals can be evaluated using a machine learning algorithm. That is, the corrected measurement signal and, optionally, the further corrected measurement signal can be used as input parameters for a machine learning algorithm. The machine learning algorithm can then provide a result for object recognition as output. Thus, the input parameter for object recognition can be adjusted according to the invention 100, or a correction parameter can be provided for this adjustment.

[0033] Further details and aspects are mentioned in connection with the exemplary embodiments described below. The in Fig. 1 The embodiment shown may include one or more optional additional features corresponding to one or more aspects related to the proposed concept or one or more embodiments described below (e.g. Fig. 2) were mentioned.

[0034] Fig. Figure 2 shows a block diagram of an embodiment of a device 30 for improving object detection in the interior of a vehicle 40 based on a measurement signal from an ultra-wideband sensor. The device 30 includes an interface 32 for communication, e.g., with an ultra-wideband sensor. The device 30 further includes a data processing circuit 34 configured to carry out at least one of the methods described herein, for example, the method which relates to Fig. 1 is described. Further embodiments include a vehicle 40 with a device 30.

[0035] The in Fig. The interface 32 shown in Figure 2 can, for example, correspond to one or more inputs and / or one or more outputs for receiving and / or transmitting information, such as digital bit values ​​based on a code, within a module, between modules, or between modules of different entities. The interface 32 can, for example, be configured to communicate with other network components via a (radio) network or a local area network.

[0036] In exemplary embodiments, the data processing circuit 34 can correspond to any controller or processor, or to a programmable hardware component. For example, the data processing circuit 34 can also be implemented as software programmed for a corresponding hardware component. In this respect, the data processing circuit 34 can be implemented as programmable hardware with appropriately adapted software. Any processor, such as digital signal processors (DSPs), can be used. These exemplary embodiments are not limited to a specific type of processor. Any processor, or even multiple processors, are conceivable for implementing the data processing circuit 34.

[0037] As in Fig. As shown in Figure 2, the interface 32 can be coupled to the respective data processing circuit 34 of the device 30. In examples, the device 30 can be implemented by one or more processing units, one or more processing devices, or any means of processing, such as a processor, a computer, or a programmable hardware component that can be operated with appropriately adapted software. Likewise, the described functions of the data processing circuit 34 can also be implemented in software, which is then executed on one or more programmable hardware components. Such hardware components can be a general-purpose processor, a digital signal processor (DSP), a microcontroller, etc.The data processing circuit 34 can be able to control the interface 32, so that any data transmission that takes place via the interface 32 and / or any interaction in which the interface 32 may be involved can be controlled by the data processing circuit 34.

[0038] In one embodiment, the device 30 may comprise a memory and at least one data processing circuit 34 which is functionally coupled to the memory and configured to perform one of the methods described above.

[0039] In examples, interface 32 can correspond to any means of receiving, receiving, transmitting, or providing analog or digital signals or information, such as any connector, contact, pin, register, input terminal, output terminal, conductor, track, etc., that enables the provision or receipt of a signal or information. Interface 32 can be wireless or wired and can be configured to communicate with other internal or external components, such as sending or receiving signals or information.

[0040] In at least some embodiments, the vehicle 40 can correspond, for example, to a land vehicle, a watercraft, an aircraft, a rail vehicle, a road vehicle, a car, a bus, a motorcycle, an all-terrain vehicle, a motor vehicle, or a truck. The device 30 can, for example, be a part of or a control unit of the vehicle 40.

[0041] Further details and aspects will be mentioned in connection with the exemplary embodiments described above. The in Fig. The embodiment shown in Figure 2 may include one or more optional additional features corresponding to one or more aspects related to the proposed concept or one or more of the above (e.g. Fig. 1) described embodiments were mentioned.

[0042] Further embodiments include computer programs for carrying out one of the methods described herein, when the computer program runs on a computer, a processor, or a programmable hardware component. Depending on specific implementation requirements, embodiments of the invention can be implemented in hardware or in software. The implementation can be carried out using a digital storage medium, for example, a floppy disk, a DVD, a Blu-ray disc, a CD, a ROM, a PROM, an EPROM, an EEPROM, a FLASH memory, a hard disk, or other magnetic or optical storage media on which electronically readable control signals are stored. These control signals can interact with, or interact with, a programmable hardware component in such a way that the respective method is carried out.

[0043] A programmable hardware component can be a processor, a computer processor (CPU = Central Processing Unit), a graphics processor (GPU = Graphics Processing Unit), a computer, a computer system, an application-specific integrated circuit (ASIC = Application-Specific Integrated Circuit), an integrated circuit (IC = Integrated Circuit), a system-on-a-chip (SOC = System on Chip), a programmable logic element, or a field-programmable gate array with a microprocessor (FPGA = Field Programmable Gate Array).

[0044] The digital storage medium can therefore be machine-readable or computer-readable. Some embodiments thus include a data carrier containing electronically readable control signals capable of interacting with a programmable computer system or a programmable hardware component to perform one of the methods described herein. An embodiment is therefore a data carrier (or a digital storage medium or a computer-readable medium) on which the program for performing one of the methods described herein is recorded.

[0045] In general, embodiments of the present invention can be implemented as a program, firmware, computer program, or computer program product with program code or as data, wherein the program code or data is / are effective in carrying out one of the methods when the program runs on a processor or a programmable hardware component. The program code or data can, for example, also be stored on a machine-readable medium or data carrier. The program code or data can be in the form of, among other things, source code, machine code, bytecode, or other intermediate code.

Claims

[1] Method (100) for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor, comprising: Obtained (110) vehicle data indicative of an interior of the vehicle; Received (120) of ultra-wideband sensor data indicative of a specification of the ultra-wideband sensor; Determining (130) a correction parameter to correct the measurement signal of the ultra-wideband sensor for object detection; Saving (140) the correction parameter to perform a correction of the measurement signal of the ultra-wideband sensor for object detection Receiving the measurement signal from the ultra-wideband sensor; Reading the correction parameter; and Correcting the measurement signal by applying the correction parameter. [2] The method (100) according to claim 1, wherein the correction comprises truncating measurement data of the measurement signal, wherein the truncated measurement data lie outside a transit time of the measurement signal determined by the correction parameter. [3] The method (100) according to any one of the preceding claims, further comprising: Receiving further ultra-wideband sensor data is indicative of a specification for another ultra-wideband sensor; Determining a further correction parameter to correct a further measurement signal from the further ultra-wideband sensor for object detection; and Saving the additional correction parameter to perform a correction of the additional measurement signal of the additional ultra-wideband sensor for object detection. [4] The method (100) according to any one of the preceding claims, further comprising: Receiving another measurement signal from the additional ultra-wideband sensor; Reading out the further correction parameter; Correcting the further measurement signal by applying the further correction parameter; and Performing object recognition based on the corrected measurement signal and the corrected further measurement signal. [5] The method (100) according to claim 4, wherein Performing object recognition includes comparing the corrected measurement signal with the corrected subsequent measurement signal; and If a discrepancy is found between the corrected measurement signal and the corrected further measurement signal during the comparison, the corrected measurement signal or the corrected further measurement signal is discarded. [6] The method (100) according to any of the preceding claims, further comprising determining a reproducible anomaly within a part of the measurement signal; and discarding the part of the measurement signal which exhibits the reproducible anomaly. [7] A computer program for carrying out one of the methods (100) according to any of the preceding claims, wherein the computer program runs on a computer, a processor, or a programmable hardware component. [8] A device (30) for improving object detection in the interior of a vehicle based on a measurement signal from an ultra-wideband sensor, comprising: an interface (32) for communication; and a data processing circuit (34) configured to carry out at least one of the methods (100) according to one of claims 1-6. [9] A vehicle (40) with a device (30) according to claim 8.

Images