Search

CN-116157312-B - Method for determining a trajectory for controlling a vehicle

CN116157312BCN 116157312 BCN116157312 BCN 116157312BCN-116157312-B

Abstract

The invention relates to a computer-implemented method for determining a trajectory (120, 122) for controlling a vehicle (100), wherein the vehicle (100) is equipped with a sensor system (102, 102a, 102 b) for capturing an environment of the vehicle (100) and a vehicle computer (112) for processing sensor data (116) and controlling the vehicle (100). The method comprises the steps of receiving sensor data (116) generated by the sensor system (102, 102a, 102 b) in a control module (206) of the vehicle computer (112), inputting the sensor data (116) into a safety algorithm, which safety algorithm has been configured to identify safety-related objects (104, 106, 109) based on the sensor data (116), inputting the sensor data (116) into a comfort algorithm, which comfort algorithm has been configured to identify comfort-related objects (108, 109) based on the sensor data (116), estimating a future state of the identified objects (104, 106, 108, 109) using an environmental model (226) representing an environment of the vehicle (100), storing the identified objects (104, 106, 108, 109) in the environmental model and tracking the identified objects over time, calculating a safety trajectory (120) taking into account a safety specification and checking whether the safety trajectory (122) meets the safety specification (122) if the safety trajectory (122) meets the safety specification (122), the safety trajectory (120) is used to control the vehicle (100).

Inventors

  • P. Shaw ter
  • G. MEYER
  • U. Colsken
  • F. Flamig
  • C. Konit

Assignees

  • 罗伯特·博世有限公司

Dates

Publication Date
20260505
Application Date
20210721
Priority Date
20200722

Claims (11)

  1. 1. A computer-implemented method for determining a trajectory (120, 122) for controlling a vehicle (100), wherein the vehicle (100) is equipped with a sensor system (102, 102a, 102 b) for capturing an environment of the vehicle (100) and a vehicle computer (112) for processing sensor data (116) and controlling the vehicle (100), wherein the method comprises: -receiving sensor data (116) generated by the sensor system (102, 102a, 102 b) in a control module (206) of the vehicle computer (112); Inputting the sensor data (116) into a security algorithm, the security algorithm having been configured to identify a security-related object based on the sensor data (116); inputting the sensor data (116) into a comfort algorithm, the comfort algorithm having been configured to identify a comfort related object based on the sensor data (116); Estimating a future state of the identified object using an environmental model (226) representing an environment of the vehicle (100), in which the identified object is stored and tracked over time; calculating a safety trajectory (120) taking into account a safety specification and a comfort trajectory (122) taking into account a comfort specification based on the estimated future state of the identified object; checking whether the comfort track (122) meets the safety specification; -if the comfort track (122) meets the safety specification, controlling the vehicle (100) using the comfort track (122); if the comfort track (122) does not meet the safety specification, the safety track (120) is used to control the vehicle (100), Wherein the output of the safety algorithm is also input into the comfort algorithm; Wherein the comfort algorithm has been configured to identify the comfort-related object based also on an output of the safety algorithm.
  2. 2. The method according to claim 1, Wherein the comfort algorithm is trained by machine learning to identify comfort-related objects and/or safety-related objects based on the sensor data (116).
  3. 3. The method according to claim 2, Wherein the comfort algorithm is based on an artificial neural network.
  4. 4. The method according to any of the preceding claims, Wherein a further safety trajectory for converting the vehicle (100) into a safety state is calculated taking into account a further safety specification based on the estimated future state of the identified object; wherein the further safety trajectory is used for controlling the vehicle (100) when the calculation of the safety trajectory (120) fails.
  5. 5. A method according to any one of claims 1 to 3, further comprising: -receiving the sensor data (116) in a further control module (208) of the vehicle computer (112), wherein the further control module (208) is supplied with electrical energy independently of the control module (206); -inputting the sensor data (116) into the security algorithm by the further control module (208); estimating, by the further control module (208), a future state of the identified object using the environmental model (226); Calculating, by the further control module (208) taking into account a further safety specification, a further safety trajectory for converting the vehicle (100) into a safety state based on the estimated future state of the identified object; Checking, by the further control module (208), whether the control module (206) is functional; The further safety trajectory (120) is used by the further control module (208) to control the vehicle (100) when the control module (206) is not active.
  6. 6. A vehicle computer (112) configured to perform the method according to any one of the preceding claims.
  7. 7. The vehicle computer (112) of claim 6, comprising: A control module (206) configured to perform the method according to any one of claims 1 to 4, and At least one further control module (208) which can be supplied with electrical energy independently of the control module (206) and which is configured to perform the method according to claim 5.
  8. 8. The vehicle computer (112) of claim 7, Wherein the control module (206) comprises a software module configured to perform the method according to any one of claims 1 to 4; Wherein the further control module (208) comprises a software module configured to perform the method of claim 5; Wherein the software modules of the further control module (208) are at least partly copies of the software modules of the control module (206).
  9. 9. A vehicle system (114), comprising: a sensor system (102, 102a, 102 b) for capturing an environment of a vehicle (100); The vehicle computer (112) of any one of claims 6 to 8, for processing sensor data (116) of the sensor system (102, 102a, 102 b) and controlling the vehicle (100).
  10. 10. A computer program product having a computer program comprising instructions which, when the computer program is executed by a vehicle computer (112) according to any one of claims 6 to 8, cause the vehicle computer (112) to perform the method according to any one of claims 1 to 5.
  11. 11. A computer readable medium having stored thereon the computer program product according to claim 10.

Description

Method for determining a trajectory for controlling a vehicle Technical Field The invention relates to a method for determining a trajectory for controlling a vehicle. Furthermore, the invention relates to a vehicle computer, a vehicle system, a computer program and a computer readable medium for performing the method. Background Autopilot places high demands on the safety of the control components of the vehicle. At the same time, a driving experience that is as pleasant as possible should be ensured. However, factors affecting the driving experience may vary greatly from one vehicle environment to another. It is therefore desirable to be able to reliably identify such factors in as many vehicle conditions as possible. The safety requirements for the control unit should continue to be met here. The vehicle may have a sensor system for capturing the vehicle environment and a vehicle computer for processing the sensor data and controlling the vehicle. The sensor system may for example comprise a plurality of different types of sensors. The vehicle computer may be configured, for example, to merge sensor data of the individual sensors, also referred to as sensor data fusion, identify objects in the vehicle environment by evaluating the merged sensor data, and calculate a suitable trajectory of the vehicle taking into account future states of both the vehicle and the identified objects. Disclosure of Invention In this context, a method, a vehicle computer, a vehicle system, a computer program and a computer-readable medium according to the invention are proposed with the solutions presented here. Advantageous developments and improvements of the solution presented here emerge from the description. THE ADVANTAGES OF THE PRESENT INVENTION Embodiments of the present invention advantageously make it possible to subdivide the architecture of a vehicle computer into a safe domain for computing as safe a trajectory as possible and a comfortable domain for computing as comfortable a trajectory as possible. The sensor data processing and/or the trajectory calculation can be performed here, for example, using different algorithms for each of the two domains. This is advantageous because the respective algorithms can be adapted specifically to the respective requirements of the domains, safety or comfort requirements, whereby on the one hand the recognition accuracy and on the other hand the computational efficiency can be increased. A first aspect of the invention relates to a computer-implemented method for determining a trajectory for controlling a vehicle, wherein the vehicle is provided with a sensor system for capturing an environment of the vehicle and a vehicle computer for processing sensor data and controlling the vehicle. The method comprises the steps of receiving sensor data generated by the sensor system in a control module of the vehicle computer, inputting the sensor data into a safety algorithm, the safety algorithm having been configured to identify safety related objects based on the sensor data, inputting the sensor data into a comfort algorithm, the comfort algorithm having been configured to identify comfort related objects based on the sensor data, estimating a future state of the identified objects using an environmental model representing an environment of the vehicle, in which the identified objects are stored and tracked over time, calculating a safety trajectory taking into account a safety specification and calculating a comfort trajectory taking into account a comfort specification, checking whether the safety specification is met by the comfort trajectory, controlling the vehicle using the comfort trajectory if the safety specification is not met by the comfort trajectory, and controlling the vehicle using the safety trajectory if the safety specification is not met by the comfort trajectory. The vehicle may be, for example, a passenger car, truck, bus or motorcycle. Alternatively, the vehicle may also be understood as a robot. The method may be automatically performed, for example, by the vehicle computer. The vehicle computer may include hardware modules and/or software modules. Thus, the control module can be implemented in hardware and/or software. In addition, the vehicle computer may include a processor, a memory, and a bus system for data communication between the processor and the memory. In addition, the vehicle computer may include one or more interfaces for data communication (also known as Car-to-X communication) with external devices (e.g., also with other vehicles or infrastructure) or with the internet. The sensor system may comprise at least one environmental sensor, such as an ultrasonic sensor, a radar sensor, a lidar sensor or a camera. Furthermore, the sensor system may comprise at least one driving dynamics sensor, such as a yaw rate sensor, an acceleration sensor, a wheel speed sensor or a steering wheel angle sensor. Furthermore, the sensor system may comprise a posit