Method, controller, and storage medium for determining an analysis and evaluation algorithm from multiple analysis and evaluation algorithms

Abstract

The present invention relates to a method for determining an analysis and evaluation algorithm (200) from a plurality of analysis and evaluation algorithms (205) for processing sensor data (210) of a vehicle sensor (110) that can be used for a vehicle (100). Herein, the method includes a reading step and a selection step. In the reading step, an environmental signal (225) is read in, and the environmental signal represents real-time environmental parameters sensed by a sensor unit having at least the vehicle sensor (110) and / or obtained via a communication interface (120). In the selection step, the analysis and evaluation algorithm (200) is selected from the plurality of analysis and evaluation algorithms (205) for analyzing and evaluating the sensor data (210) of the vehicle sensor (110) by using the environmental parameters. The present invention also relates to a controller, a computer program, and a machine-readable storage medium.

Description

Technical Field

[0001] The present invention starts from a method and a controller for determining one analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for processing sensor data of vehicle sensors that can be used in a vehicle. The subject matter of the present invention also relates to a computer program. Background Art

[0002] An autonomous vehicle is a vehicle that can also cope without a driver. Here, the vehicle travels autonomously in such a way that the vehicle independently identifies, for example, the road direction, other traffic participants or obstacles, calculates corresponding control instructions in the vehicle, and forwards the control instructions to the actuators in the vehicle, thereby correctly influencing the driving direction of the vehicle. The driver does not participate in the driving process in a fully autonomous vehicle. Currently available vehicles cannot yet act autonomously. On the one hand, because the corresponding technology is not yet fully mature. On the other hand, because currently it is still legally stipulated that the person guiding the vehicle must be able to intervene in the driving process at any time by himself. This makes it difficult to implement autonomous vehicles. However, there are already systems that embody highly automated driving. Therefore, it can now be foreseen that fully autonomous vehicle systems will appear on the market in a few years. Summary of the Invention

[0003] Against this background, with the method proposed here, an improved method for determining one analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for processing sensor data of vehicle sensors, an improved controller using this method, and finally a corresponding computer program are proposed. Advantageous extensions and improvements of the controller of the present invention can be achieved by the measures listed in the preferred embodiments.

[0004] The method proposed here enables the possibility of processing sensor data as well as possible, for example to improve functionality and thus increase reliability and safety.

[0005] A method for determining one analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for processing sensor data of vehicle sensors is proposed. Here, the method includes a step of reading and a step of selection. In the step of reading, such environmental signals are read, which represent real-time environmental parameters sensed by a sensor unit having at least vehicle sensors and additionally or alternatively represent real-time environmental parameters obtained via a communication interface. In the step of selection, one analysis and evaluation algorithm is selected from a plurality of analysis and evaluation algorithms for analyzing and evaluating the sensor data of vehicle sensors using the environmental parameters.

[0006] This method can be used, for example, for vehicles configured to transport people and additionally or alternatively transport objects. The vehicle can be, for example, a highly automated vehicle correspondingly having at least one vehicle sensor. The vehicle sensor is configured to be able to advantageously implement safety-related functions of the vehicle. The vehicle sensor can be implemented, for example, as an optical sensor such as a camera or a lidar sensor. Alternatively, the vehicle sensor can also be implemented as a radar or ultrasonic sensor. For example, at least one analysis and evaluation algorithm selected from a plurality of analysis and evaluation algorithms can be used to process sensor data sensed by the vehicle sensor, such as image data. This means that each of these analysis and evaluation algorithms includes determination conditions for one analysis and evaluation. At least one analysis and evaluation algorithm can be implemented, for example, as a neural network or artificial intelligence. Real-time environmental parameters sensed by the vehicle sensor or obtained via a communication interface can, for example, display environmental information at a certain time point at the current position of the vehicle. Advantageously, the safety of at least one occupant of the vehicle can be ensured by this method.

[0007] According to one embodiment, in the reading step, an environmental signal representing environmental parameters can be read, where the environmental parameters represent driving parameters, weather parameters, traffic parameters, and additionally or alternatively sensor-specific environmental parameters. The environmental parameter can be, for example, a value in the form of a driving parameter such as the vehicle speed or tire pressure that may be related to the vehicle. The driving parameter can advantageously represent such a value that relates to the vehicle and additionally or alternatively relates to the driving of the vehicle. For example, such values related to air pressure, air flow, and various weather conditions can be referred to as weather parameters. Thereby, it is possible to advantageously distinguish between rainy weather with abundant rainfall and good, sunny weather. Traffic parameters represent values representing, for example, traffic density and other traffic-related data. For example, data preprocessed by other sensors of the vehicle can be referred to as sensor-specific environmental parameters. Advantageously, this can provide as comprehensive a guarantee as possible for the vehicle and additionally or alternatively for vehicle functions.

[0008] In the selection step, one analysis and evaluation algorithm can be selected from a plurality of analysis and evaluation algorithms, where at least one environmental parameter assigned to the selected analysis and evaluation algorithm corresponds to the current environmental parameter. This means that, for example, the current environmental parameter is compared with the environmental parameters stored in each analysis and evaluation algorithm and then an analysis and evaluation algorithm is selected such that in this analysis and evaluation algorithm, the stored environmental parameter corresponds to the current environmental parameter, for example, the current driving parameter, the current weather parameter, the current traffic parameter, and additionally or alternatively corresponds to the current sensor-specific environmental parameter, for example, within a tolerance range. Advantageously, accurate and reliable results for the corresponding situation can be obtained by applying the selected analysis and evaluation algorithm.

[0009] According to one embodiment, in the selection step, a second analysis and evaluation algorithm can be selected from a plurality of analysis and evaluation algorithms for analyzing and evaluating the sensor data of a vehicle by using environmental signals and additionally or alternatively by using the analysis and evaluation results, where the analysis and evaluation results represent the results of applying the selected analysis and evaluation algorithm to the sensor data. This means that the second analysis and evaluation algorithm is also selected by using environmental parameters, so that the second analysis and evaluation algorithm can also produce reliable results. Advantageously, it can be ensured that objects are recognized as well as possible in the environment of the vehicle.

[0010] According to another embodiment, in the selection step, the second analysis and evaluation algorithm can be selected by using environmental parameters of the environmental signal that are different from the environmental parameters of the environmental signal used for selecting the analysis and evaluation algorithm. This means that, for example, the analysis and evaluation algorithm and the second analysis and evaluation algorithm can be executed by using different environmental parameters, such as high speed or low speed and good weather or bad weather as physical parameters, so that advantageously, the results will not be distorted due to biases of factors such as these.

[0011] Furthermore, the method can include steps of applying the analysis and evaluation algorithm and the second analysis and evaluation algorithm, where the analysis and evaluation algorithm and the second analysis and evaluation algorithm can be implemented at least partially simultaneously and additionally or alternatively in sequence. This means that these two analysis and evaluation algorithms can be implemented, for example, not only simultaneously but also successively with each other. Thus, for example, while one of these analysis and evaluation algorithms is activated or deactivated in the background, the other analysis and evaluation algorithm can be activated in the foreground. Advantageously, dynamic activation and additionally or alternatively deactivation can be achieved thereby.

[0012] According to one embodiment, in the application step, the analysis and evaluation results of applying the selected analysis and evaluation algorithm to the sensor data and the second analysis and evaluation results of the second analysis and evaluation algorithm applied at least partially simultaneously with respect to the selected analysis and evaluation algorithm can be cached. Here, in response to a switching signal, the output can be switched from the analysis and evaluation results to the second analysis and evaluation results. This means that, for example, the analysis and evaluation results and the second analysis and evaluation results can be stored in, for example, a circular memory, and for example, even after switching the analysis and evaluation algorithm activated in the foreground, the analysis and evaluation results of the analysis and evaluation algorithm activated in the foreground are always stored. Advantageously, the analysis and evaluation results and the additional or alternative second analysis and evaluation results can be directed to further applications of the analysis and evaluation algorithm.

[0013] In the application step, the selected analysis and evaluation algorithm can be applied by using the same environmental parameters of the environmental signal as those of the second analysis and evaluation algorithm. Thereby, the analysis and evaluation results can be advantageously checked.

[0014] According to one embodiment, in the application step, the selected analysis and evaluation algorithm and the second analysis and evaluation algorithm can be fed with the same sensor data. Advantageously, in this way, the two analysis and evaluation results can be assigned to a specific point in time, for example.

[0015] According to one implementation method, in the reading-in step, at least one of the analysis and evaluation algorithms can be read in from a vehicle external device, the cloud, and an additional or alternative other vehicle, and additionally or alternatively, at least one selected analysis and evaluation algorithm can be provided to the vehicle external device, the cloud, and additionally or alternatively to another vehicle in the selection step. The vehicle external device can be, for example, a computer or a traffic device outside the vehicle, through which traffic data can be cached and transmitted to other vehicles, for example. Alternatively, the vehicle can communicate directly with another vehicle so that the corresponding data can be directly forwarded. The cloud can represent, for example, a virtual storage space in which the data output by the vehicle can be cached. Advantageously, traffic safety can be further improved by transmitting the data.

[0016] The method can be implemented, for example, in software or hardware or in a hybrid form of software and hardware, such as in a controller.

[0017] The method proposed here also implements a controller that is configured to implement, control, or carry out the steps of a variant of the method proposed here in the corresponding device. These variant embodiments of the present invention in the form of a controller can also quickly and effectively solve the problem underlying the present invention.

[0018] For this purpose, the controller can have at least one computing unit for processing signals or data, at least one storage unit for storing signals or data, at least one interface to sensors for reading in sensor signals or at least one interface to actuators for outputting control signals, and / or at least one communication interface for reading in or outputting data embedded in a communication protocol. The computing unit can be, for example, a signal processor, a microcontroller, etc., and the storage unit can be a flash memory, an EEPROM, or a magnetic storage unit. The communication interface can be configured to read in or output data wirelessly and / or wiredly, and a communication interface that can read in or output wired data can read in data from or output the data to a corresponding data transmission line, for example, in an electrical or optical manner.

[0019] Here, the controller can be understood as an electrical device that processes sensor signals and outputs control signals and / or data signals based on the sensor signals. The controller can have an interface that can be constructed in hardware form and / or in software form. In the case of being constructed in hardware form, the interface can be, for example, part of a so-called system ASIC, and this part of the system ASIC contains various functions of the controller. However, it is also feasible that the interface is an integrated circuit of its own, or the interface consists at least partially of discrete components. In the case of being constructed in software form, the interface can be, for example, a software module that exists on a microcontroller together with other software modules.

[0020] In an advantageous configuration, a method for determining an analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms available for processing sensor data of vehicle sensors is controlled by the controller. For this purpose, the controller can, for example, call a sensor signal, such as an environmental signal, which represents real-time environmental parameters sensed by a sensor unit having at least vehicle sensors and / or obtained via a communication interface. Manipulation is performed by an actuator, such as a reading unit and a selection unit. The reading unit is configured to read in the environmental signal, and the selection unit is configured to select an analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for analyzing and evaluating the sensor data of the vehicle sensors using the environmental signal.

[0021] A computer program product or a computer program having program code is also advantageous. The program code can be stored on a machine-readable carrier or storage medium, such as a semiconductor memory, a hard disk, or an optical memory, and is used to implement, realize, and / or control the steps of the method according to one of the foregoing embodiments, especially when the program product or program is implemented on a computer or a controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Embodiments of the method proposed here are shown in the drawings and elaborated in detail in the following description. The drawings show:

[0023] Figure 1 A schematic diagram of a vehicle having a controller according to an embodiment;

[0024] Figure 2 A schematic diagram of a controller according to an embodiment;

[0025] Figure 3 A schematic diagram of a controller according to an embodiment;

[0026] Figure 4 A flowchart of a method for determining an analysis and evaluation algorithm from a plurality of available analysis and evaluation algorithms according to an embodiment.

[0027] In the following description of advantageous embodiments of the present invention, elements that are shown in different figures and have similar functions are provided with the same or similar reference signs, and the repeated description of these elements is omitted. Detailed Description

[0028] Figure 1 Schematic illustration of a vehicle 100 having a controller 105 according to an embodiment. Here, the vehicle 100 is configured to transport persons and / or objects and is configured as a highly automated vehicle 100 according to this embodiment. According to this embodiment, the controller 105 is configured to implement or control a method for determining one analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for processing sensor data of vehicle sensors 110 available for the vehicle 100. Furthermore, the vehicle 100 has vehicle sensors 110 which, according to this embodiment, are part of a sensor unit which can, for example, have a plurality of sensors, such as optical sensors. Furthermore, the vehicle 100 has a storage device 115 according to this embodiment, which is configured to store, for example, a plurality of analysis and evaluation algorithms. Alternatively, other data, such as sensor data sensed by the vehicle sensors 110, can also be stored, which sensor data is, for example, further processed at a later point in time. The storage device 115 can, for example, be implemented as a rewritable ring memory. The vehicle 100 also has a communication interface 120 which is configured to communicate wirelessly, for example, with an external device, a cloud or another vehicle. Alternatively, the storage device can also be configured as part of the controller 105 or integrated into the controller.

[0029] In other words, a plurality of vehicle sensors 110, which can also be referred to jointly as a sensor unit, are used in highly automated and fully automated vehicles 100, also referred to as autonomous vehicles. According to this embodiment, such a vehicle 100 can communicate, for example, via vehicle-to-X communication with devices outside the vehicle or other vehicles. This means that information and data are exchanged between these motor vehicles and / or between the motor vehicle and the surrounding infrastructure, such as traffic lights. The aim is to inform the driver early of critical and / or dangerous situations. For example, the vehicle 100 collects data, such as ABS interventions, steering angles, positions, directions and speeds, and sends this data to other road users or traffic infrastructure via radio, such as WLAN or UMTS, for example by means of the communication interface 120.

[0030] In addition, a vehicle 100, also referred to as a motor vehicle, has a driver assistance system. The driver assistance system is implemented, for example, as an electronic add-on in the vehicle 100 to assist the driver in certain driving situations. Here, aspects of safety are usually important, but driving comfort is also important. On the other hand, there is an improvement in economy. The driver assistance system intervenes partially autonomously and / or autonomously in the drive device, control devices such as the throttle or brakes, or partially autonomously and / or autonomously in the signal emitting devices of the vehicle 100, or the driver assistance system warns the driver via a suitable man-machine interface in the event of an impending emergency or during an emergency. Currently, most driver assistance systems are designed such that responsibility remains with the driver and the driver is not relieved of this responsibility. Different types of environmental sensor devices are used for the driver assistance system, such as ultrasonic (parking assistance), radar (lane change assistance, automatic distance warning), lidar (blind spot monitoring, automatic distance warning, distance regulation, "Pre-Crash" and "Pre-Brake"), camera (lane departure warning, traffic sign recognition, lane change assistance, blind spot monitoring, emergency braking system for pedestrian protection) and / or GNSS (high-precision vehicle positioning on the map, control of autonomous vehicles, Safe-Stop). Here, GNSS stands for Global Navigation Satellite System (English) by receiving signals from navigation satellites on and / or above the earth.

[0031] According to this embodiment, the vehicle sensor 110 or, alternatively, a controller downstream thereof also has an analysis and evaluation algorithm to process the data of the vehicle sensor 110. For example, objects around the vehicle 100 are identified based on sensor data on the vehicle sensor 110 by means of a neural network or artificial intelligence (KI). In addition, the sensor data can be pre-filtered by means of a corresponding analysis and evaluation algorithm, also referred to as an algorithm, or the sensor data can be pre-processed for subsequent processing. In addition to object recognition, trajectory planning is also carried out on the vehicle 100 based on the instantaneous, meaning real-time, environmental sensor data, also referred to here as environmental parameters, and the objects identified therein. Such an analysis and evaluation algorithm or such artificial intelligence has limited functional capabilities under certain conditions. For this reason, the controller 105 proposed here is advantageously configured to select one analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms.

[0032] Figure 2 A schematic diagram of a controller 105 according to an embodiment is shown. The controller can be the controller 105 described in Figure 1 as described. The controller 105 can, for example, be inFigure 1 It is used in the vehicle described in Figure 1 . The controller 105 is configured to implement a method for determining the analysis and evaluation algorithm 200 from a plurality of analysis and evaluation algorithms 205 for processing the sensor data 210 of the vehicle sensors 110 available for the vehicle. To this end, the controller 105 has an input unit 215 and a selection unit 220. The input unit 215 is configured to input an environmental signal 225. The environmental signal 225 represents real-time environmental parameters sensed by a sensor unit having at least one vehicle sensor 110 and / or obtained via a communication interface 120. Here, according to this embodiment, the environmental parameters represent driving parameters, weather parameters, traffic parameters, and / or sensor-specific environmental parameters. It should be noted that the sensors for the sensor data 210 can be different from the sensors for the environmental signal 225. However, it is also conceivable that the sensor data 210 and the environmental signal 225 are provided by the same sensor.

[0033] The selection unit 220 is configured to select the analysis and evaluation algorithm 200 from a plurality of analysis and evaluation algorithms 205 so that the sensor data 210 of the vehicle sensors 110 can be analyzed and evaluated using the environmental parameters, and according to this embodiment, an analysis and evaluation result 230 is obtained. Thus, the analysis and evaluation result 230 can be used, for example, for further processing. Here, according to this embodiment, the environmental parameters assigned to the selected analysis and evaluation algorithm 200 correspond to the current environmental parameters.

[0034] In other words, the possibility of executing multiple analytical evaluation algorithms 205 for determined environmental parameters is proposed. For example, in order to analyze and evaluate the sensor data 210 of a lidar sensor in good weather, a specific analytical evaluation algorithm 200 or a specific neural network or a specific artificial intelligence for processing exactly these environmental sensor data under exactly these weather conditions is used. In case of bad weather or for example in case of snowfall, according to this embodiment, a second analytical evaluation algorithm 235 is used to process the sensor data 210 of the vehicle sensor 110. This means that a switch between different analytical evaluation algorithms 200, 235 is made based on certain input data, such as weather conditions. This has obvious advantages: for example, different analytical evaluation algorithms 200, 235 are used to process the data of the corresponding vehicle sensor 110 in the vehicle for each weather condition (sunshine, rain, snow, fog, dust, etc.) or for each additional environmental condition or input condition read in via the environmental signal 225. A switch between different analytical evaluation algorithms 200, 235 is made when the environmental parameters, which can also be referred to as input conditions, change, so that at each point in time, the vehicle sensor 110 in the vehicle has the greatest possible performance available. In this way, highly automated driving is configured more safely. According to this embodiment, a switch between different analytical evaluation algorithms 200, 235 is made with the aid of input data called environmental parameters, which input data relate, for example, to weather conditions, vehicle speed, traffic density and sensor-specific environmental parameters.

[0035] Depending on the different weather conditions, it may be necessary to switch to the analytical evaluation algorithm 200. For example, in case of good weather conditions, the data of the lidar sensor are analyzed and evaluated differently than in case of snowfall or rainfall. For this purpose, according to this embodiment, different analytical evaluation algorithms 200, 235 are stored on the controller 105, and these analytical evaluation algorithms are switched on or off depending on the weather conditions.

[0036] Alternatively, the analysis and evaluation algorithms 200, 235 are specifically designed for different environmental parameters. This means that on the vehicle sensor 110 or the downstream processing unit (controller) 105, exactly only one analysis and evaluation algorithm 200, 235 for a determined weather condition, determined vehicle speed, determined traffic density, etc. always runs at each time point. This means that according to one embodiment, multiple analysis and evaluation algorithms 200, 235 run sequentially on the vehicle sensor 110 and / or on the controller 105. In this case, the analysis and evaluation algorithms 200, 235 are designed such that as many environmental conditions as possible are satisfied with as large a performance as possible by means of a small number of analysis and evaluation algorithms 200, 235. The environmental conditions can for example correspond to a weather condition algorithm for good weather, a weather condition algorithm for bad weather, a speed algorithm at low speed, a speed algorithm at high speed, a weather condition and speed algorithm in the case of good weather and low speed, a weather condition and speed algorithm for good weather at high speed, a weather condition and speed algorithm in the case of bad weather and low speed, and a weather condition and speed algorithm in the case of bad weather and high speed. This is only a simplified example according to this embodiment. Here, the algorithms 200, 235 cover at least one input variable, such as the weather condition. A simple switch based only on, for example, the weather condition without a combination of different input data is also feasible in this embodiment. The number and type of different analysis and evaluation algorithms 200, 235 based on different input conditions can for example depend on the vehicle sensor type. For example, for a lidar sensor, it makes sense to switch the analysis and evaluation algorithms 200, 235 based on the weather, while for example for a radar sensor, it makes sense to switch according to the traffic density. For example, for a camera sensor, it may be necessary to switch according to the speed in order to respectively identify objects in the image stream at different speeds with good performance, because for example at higher speeds, the objects will be more distorted.

[0037] According to one embodiment, the analysis and evaluation algorithms 200, 235 are designed for different input data such that these analysis and evaluation algorithms can for example be successively implemented in a cascaded manner. For example, the sensor data 210 is preprocessed by means of the mentioned analysis and evaluation algorithms 200, 235 according to, for example, the weather condition, and subsequently the sensor data 210 is further processed by a downstream analysis and evaluation algorithm, which is referred to as the second analysis and evaluation algorithm 235 according to this embodiment, according to, for example, the vehicle speed or the traffic density. Thereby, the number of the saved analysis and evaluation algorithms 200, 235 is advantageously reduced such that not every single input condition must have its own algorithm, and thereby, the storage amount in the vehicle is also saved. Here, the analysis and evaluation algorithms 200, 235, which are also referred to as algorithms within the cascade, can generate output data that is as good as possible according to different input data, which means producing analysis and evaluation results.

[0038] For example, according to this embodiment, the analysis and evaluation algorithms 200, 235 for analyzing and evaluating environmental sensor data related to vehicle speed can be designed, trained, or learned with the help of the output data of the upstream algorithm for weather conditions. Next, the corresponding vehicle speed algorithm can perform further processing on the sensor data 210 based on the results of the weather condition algorithm, based on the corresponding different analysis and evaluation algorithms 200, 235 for weather conditions that are located before the vehicle speed algorithm in the cascade. This means, for example, switching the weather condition algorithm in the case of a change in weather conditions, but not necessarily also switching the algorithm for vehicle speed analysis and evaluation. This has the obvious advantage that fewer algorithms have to be stored. According to this embodiment, examples thereof are the weather condition algorithm for good weather and the vehicle speed algorithm at low speed, the weather condition algorithm for good weather and the vehicle speed algorithm at high speed, the weather condition algorithm for bad weather and the vehicle speed algorithm at low speed, and the weather condition algorithm for bad weather and the vehicle speed algorithm at high speed. This is only a simplified example. Here, the cascade of algorithms includes at least two algorithms for different input conditions and these algorithms can be of any length, resulting in different permutation possibilities.

[0039] According to one embodiment, the analysis and evaluation algorithms 200, 235 are intelligent algorithms, such as neural networks or artificial intelligence (AI). These intelligent algorithms are learned, for example, with the help of training data for different weather conditions or different vehicle speeds, or depending on the implementation, using a combination of the corresponding relevant input data. Next, the corresponding algorithms 200, 235 are stored in the vehicle sensors 110 and / or the controller 105 during production. It is also conceivable that, according to one embodiment, the corresponding algorithms 200, 235 are learned in the cloud and then the algorithms are transmitted to the corresponding vehicle and its vehicle sensors 110 and / or the controller 105 via a vehicle-to-X communication connection.

[0040] Instead of or in addition to the scenario shown in Figure 2 , the analysis and evaluation algorithm 200 or multiple analysis and evaluation algorithms 205 can also be arranged or stored on one of the vehicle sensors 110 without the need to be integrated into a separate controller. In this case, for example, the controller 105 can be understood as part of one of the vehicle sensors 110. It is also conceivable that only one or multiple of the analysis and evaluation algorithms 200 are stored on one or multiple of these vehicle sensors 110 and are "downloaded" accordingly by the (e.g., separate) controller 105 when selected accordingly. In this way, for example, corresponding different analysis and evaluation algorithms can also be set for each corresponding sensor type, and these analysis and evaluation algorithms are used by the (e.g., central) controller 105.

[0041] Alternatively or additionally, the analysis and evaluation algorithm 200 can also be read in via the communication interface 120. This communication interface 120, which can be implemented, for example, as a mobile radio connection such as a UMTS connection, an EDGE connection, an LTE connection, a 5G connection, a WLAN connection or a Bluetooth connection or a similar connection, enables the controller 105, for example, to load one or more analysis and evaluation algorithms 200 from a central server or the cloud and use the analysis and evaluation algorithms accordingly in the controller 105. Such an embodiment has the advantage of a quick configuration of the provided analysis and evaluation algorithms, so that, for example, even after the delivery of the controller 105 or the vehicle sensors 110, the analysis and evaluation algorithm 200 can still be adapted or optimized. In principle, the selection of the analysis and evaluation algorithm can also be carried out in a central server, for example in the cloud. For this purpose, the environmental signal 225 (also in the case of using the communication interface 120, for example) should be transmitted to this central server or the cloud, so that the required information or available analysis and evaluation algorithms are available in the cloud or the central server in order to be able to select the most advantageous or most suitable analysis and evaluation algorithm. Thus, it can be seen that the method proposed here can be implemented not only in the units of the vehicle 100, but the method proposed here can be implemented at any location inside or outside the vehicle. Thus, an embodiment can be implemented in which the relevant or selected analysis and evaluation algorithm 200 is loaded from a certain amount of analysis and evaluation algorithms located in the cloud and the analysis and evaluation algorithm is then run on the selection unit 220, or, if the selection unit is implemented in the cloud, the selected analysis and evaluation algorithm is loaded and run in the controller 105.

[0042] Figure 3 Schematic illustration of a controller 105 according to an embodiment. The controller 105 shown here can be an alternative embodiment of the controller 105 described in Figure 2 and can be implemented for a vehicle as described in Figure 1 According to this embodiment, the controller 105 also has an input unit 215 and a selection unit 220 as described in Figure 2 Here, the input unit 215 is also configured to input the environmental signal 225. The selection unit 220 is configured to select the analysis and evaluation algorithm 200. Additionally, according to this embodiment, the selection unit 220 is configured to select a second analysis and evaluation algorithm 235 for analyzing the sensor data 210 in the case of using the environmental signal 225 (which means, for example, in the case of using environmental parameters of the environmental signal 225 that are the same or different compared to the environmental parameters of the environmental signal 225 used for selecting the analysis and evaluation algorithm 200) and / or according to this embodiment in the case of using the analysis and evaluation result 305. According to this embodiment, the analysis and evaluation algorithm 200 and the second analysis and evaluation algorithm 235 can be implemented at least partly simultaneously and / or sequentially here.

[0043] The analysis and evaluation result 305 represents, for example, the result of applying the selected analysis and evaluation algorithm 200 to the sensor data 210. Here, the second analysis and evaluation result 310 represents, for example, the second result of applying the selected second analysis and evaluation algorithm 235 to the sensor data 210. After switching from outputting the analysis and evaluation result 305 to outputting the second analysis and evaluation result 310 or from outputting the second analysis and evaluation result 310 to outputting the analysis and evaluation result 305 in response to the switching signal 325 according to this embodiment, according to this embodiment, the analysis and evaluation result 305 and the second analysis and evaluation result 310 are cached, for example, on the storage device 315 by means of the storage signal 320. Here, the storage device 315 is implemented as or can be implemented as a circular memory, for example.

[0044] In other words, according to this embodiment, the controller 105 reads in the corresponding input data for switching the saved analysis and evaluation algorithms 200, 235 and then releases or starts the corresponding analysis and evaluation algorithms 200, 235. This can be done dynamically, for example, during vehicle operation. For this purpose, for example, the determined sensor data 210 of the vehicle sensors are written into the storage device 315, such as a cyclic circular memory, and these sensor data are analyzed and evaluated by means of the analysis and evaluation algorithm 200 activated for one input condition. In the background, the second analysis and evaluation algorithm 235 to be switched to immediately is also started and / or run by means of the cached sensor data 210. Next, the still active algorithm 200 is deactivated and quickly switched to the now stable second algorithm 235 in such a way that from now on, only the data of the switched-on second algorithm 235 are used for driving, and no longer the data of the deactivated analysis and evaluation algorithm 200. The calculation based on the deactivated analysis and evaluation algorithm 200 is interrupted and the circular memory is only analyzed and evaluated with the second algorithm 235. In this way, the analysis and evaluation algorithms 200, 235 are switched in a safe manner by first calculating the two algorithms 200, 235 in parallel and interrupting the first algorithm 200 only when the second algorithm 235 is in a stable state. Optionally, for example, weather data are obtained from a weather server outside the vehicle or in the cloud according to the instantaneous vehicle position and the weather data are transmitted into the vehicle, for example, via a vehicle-to-X communication connection. According to one embodiment, the instantaneous vehicle speed is obtained, for example, by means of a GNSS-based sensor in the vehicle or by means of wheel speed sensor data. According to one embodiment, optionally, for example, traffic density is also received from the cloud, which receives the corresponding vehicle positions in a determined area over time and combines the vehicle positions with traffic density information. In addition, according to one embodiment, the traffic density is determined by means of environmental sensors and object recognition installed in the vehicle.

[0045] Figure 4A flowchart of a method 400 for determining an analysis and evaluation algorithm from a plurality of available analysis and evaluation algorithms according to an embodiment is shown. According to this embodiment, as illustrated in Figures 1 to 3 , the method 400 described herein can be implemented by a controller. Here, the method 400 includes a step of reading in 405 environmental signals, which represent real-time environmental parameters sensed by a sensor unit having at least vehicle sensors and / or obtained via a communication interface. In addition, the method 400 includes a step of selecting 410 an analysis and evaluation algorithm from a plurality of analysis and evaluation algorithms for analyzing and evaluating sensor data of vehicle sensors using the environmental parameters. Optionally, in the step of reading in 405, at least one of the analysis and evaluation algorithms is read in from an external vehicle device, the cloud, and / or another vehicle, and / or in the step of selecting 410, at least one selected analysis and evaluation algorithm is provided to the external vehicle device, the cloud, and / or the other vehicle.

[0046] In addition, the method 400 according to this embodiment includes a step of applying 415 the analysis and evaluation algorithm and a second analysis and evaluation algorithm, wherein the analysis and evaluation algorithm and the second analysis and evaluation algorithm are executed at least partially simultaneously and / or sequentially. Here, in the step of applying 415, the selected analysis and evaluation algorithm is applied using environmental parameters of the same environmental signal as the second analysis and evaluation algorithm. According to this embodiment, the selected analysis and evaluation algorithm and the second analysis and evaluation algorithm are optionally fed the same sensor data.

[0047] If an embodiment includes an "and / or" association between a first feature and a second feature, this should be interpreted as follows: The embodiment has both the first feature and the second feature according to one implementation, and according to another implementation, has either only the first feature or only the second feature.

Claims

1. A method (400) for determining an analysis and evaluation algorithm (200) from a plurality of analysis and evaluation algorithms (205) for sensor data (210) of a vehicle sensor (110) of a vehicle (100) available for processing automation, wherein, The method (400) comprises the following steps: - Reading in (405) an environmental signal (225) representing real-time environmental parameters sensed by a sensor unit having at least the vehicle sensor (110) and / or obtained via a communication interface (120); and - Selecting (410) the analysis and evaluation algorithm (200) from the plurality of analysis and evaluation algorithms (205) for analyzing and evaluating the sensor data (210) of the vehicle sensor (110) using the environmental parameters, wherein at least one environmental parameter assigned to the selected analysis and evaluation algorithm (200) corresponds to the current environmental parameter. wherein, in the step of selection (410), a second analysis and evaluation algorithm (235) for analyzing and evaluating the sensor data (210) of the vehicle sensor (110) is selected from the plurality of analysis and evaluation algorithms (205) using the environmental signal (225) and / or using the analysis and evaluation result (305), the analysis and evaluation result representing the result of applying the selected analysis and evaluation algorithm (200) to the sensor data (210); wherein the vehicle sensor (110) is part of a sensor unit having a plurality of sensors; wherein the vehicle (100) has a storage device (115) for storing a plurality of analysis and evaluation algorithms; wherein the vehicle (100) further has a communication interface (120) configured for wireless communication with an external device, a cloud, and / or other vehicles; wherein each analysis and evaluation algorithm (200) in the plurality of analysis and evaluation algorithms (205) corresponds to a specific environmental parameter, and the analysis and evaluation algorithm (200) is selected from the plurality of analysis and evaluation algorithms (205) in such a way that the environmental parameter assigned to the selected analysis and evaluation algorithm (200) corresponds to the specific environmental parameter; wherein a switch between different analysis and evaluation algorithms is made based on changing weather conditions and / or other changing conditions; A switch between different analysis and evaluation algorithms is made when the environmental parameters change, such that at each time point, as much performance as possible of the vehicle sensor (110) in the vehicle is available, wherein the switch between different analysis and evaluation algorithms is made using environmental parameters related to weather conditions, vehicle speed, traffic density, and sensor-specific environmental parameters. The environmental parameters represent driving parameters, weather parameters, traffic parameters, and / or sensor-specific environmental parameters.

2. The method (400) according to claim 1, wherein, in the method, an environmental signal (225) representing the environmental parameter is read in (405) in the step of reading in, where 3. The method (400) according to claim 1, wherein in the step of selection (410), the second analysis and evaluation algorithm (235) is selected using an environmental parameter of the environmental signal (225) that is different from the environmental parameter of the environmental signal (225) used for selecting the analysis and evaluation algorithm (200). ​ 4. The method (400) according to claim 1, the method having the step of applying (415) the analysis and evaluation algorithm (200) and the second analysis and evaluation algorithm (235), wherein, Implement the analysis and evaluation algorithm (200) and the second analysis and evaluation algorithm (235) at least partially simultaneously and / or sequentially.

5. The method (400) according to claim 4, wherein in the step of application (415), the analysis and evaluation result (305) obtained by applying the selected analysis and evaluation algorithm (200) to the sensor data (210) and the second analysis and evaluation result (310) of a second analysis and evaluation algorithm (235) applied at least partially simultaneously with respect to the selected analysis and evaluation algorithm (200) are cached, where Switch from outputting the analysis and evaluation result (305) to outputting the second analysis and evaluation result (310) in response to a switching signal (325).

6. The method (400) according to claim 5, in which, in the step of application (415), the analysis and evaluation algorithm (200) selected in the case of using the same environmental parameters of the environmental signal (225) as those of the second analysis and evaluation algorithm (235) is applied.

7. The method (400) according to claim 5 or 6, wherein In the step of application (415), the selected analysis and evaluation algorithm (200) and the second analysis and evaluation algorithm (235) are fed with the same sensor data (210).

8. The method (400) according to any one of claims 1 to 6, wherein In the step of reading in (405), at least one of the analysis and evaluation algorithms (200, 235) is read in from a vehicle external device, the cloud and / or another vehicle via the communication interface (120), and / or, in the step of selection (410), at least one selected analysis and evaluation algorithm (200, 235) is provided to the vehicle external device, the cloud and / or another vehicle via the communication interface (120).

9. A controller (105), the controller being configured to implement and / or control the steps of the method (400) according to any one of claims 1 to 8 in corresponding units (215, 220).

10. A computer program product, the computer program product being configured to implement and / or control the steps of the method (400) according to any one of claims 1 to 8.

11. A computer-readable storage medium, on which the computer program product according to claim 10 is stored.

Images