CN-116443047-B - Vehicle control method, device, equipment and storage medium

CN116443047BCN 116443047 BCN116443047 BCN 116443047BCN-116443047-B

Abstract

The application discloses a vehicle control method, a device, equipment and a storage medium. And acquiring space-time partitions corresponding to at least two preset driving decisions respectively. And determining decision driving data corresponding to the decision space range of the driving decision from the driving data determined in the acquired decision data acquisition period corresponding to the driving decision when the decision period corresponding to the driving decision is reached for each driving decision. And finally, planning and executing a driving instruction corresponding to the driving decision according to the decision driving data. According to the embodiment of the application, the vehicle is controlled to run according to the determined driving instruction. The driving instruction is obtained by planning the decision driving data corresponding to the space-time partition of the driving decision, which is acquired in the decision data acquisition period corresponding to the driving decision, so that the data acquired in each decision data acquisition period is effectively utilized, and the efficiency is improved.

Inventors

ZHANG LIN
ZHANG WENDOU
GUO XIAOYING
CAO LI

Assignees

北京经纬恒润科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20221228

Claims (10)

1. A vehicle control method characterized by comprising: Acquiring a space-time partition corresponding to at least two preset driving decisions respectively, wherein the space-time partition at least comprises a decision data acquisition period, a decision period and a decision space range, For each driving decision, when the decision period corresponding to the driving decision is reached, obtaining the driving data determined in the decision data acquisition period corresponding to the driving decision which is finished, Determining decision driving data corresponding to a decision space range of the driving decision from the driving data, Planning a driving instruction corresponding to the driving decision according to the decision driving data, Controlling the vehicle to run according to the driving instruction; acquiring a decision data acquisition cycle comprising: For each driving decision, determining at least one target vehicle sensor corresponding to the driving decision, Respectively acquiring data acquisition periods of the at least one target vehicle sensor, Determining the maximum data acquisition period in at least one data acquisition period as a decision data acquisition period; Obtaining driving data determined in a decision data acquisition period corresponding to the driving decision after finishing, including: Sensor data acquired by a vehicle sensor during a decision data acquisition period corresponding to the driving decision that has ended and for which the time interval at the present moment is the smallest is acquired, Fitting the sensor data by a linear interpolation algorithm according to the sensor data acquired by the vehicle sensor in a decision data acquisition period corresponding to the finished driving decision with the minimum time interval at the current moment to obtain fitted sensor data corresponding to the current moment, Determining the obtained fitting sensor data as driving data; according to the decision driving data, planning a driving instruction corresponding to the driving decision, including: when the driving decision is a lane-level driving decision, determining a distance to be driven of a non-driven part of the predicted driving distances of the driving instructions according to the decision driving data and the driving instructions being executed, When the distance to be driven is determined to be larger than a preset distance threshold value, determining the driving instruction as a re-planned driving instruction, And when the distance to be driven is determined to be smaller than or equal to the distance threshold value, re-planning and executing a driving instruction according to decision driving data corresponding to the lane-level driving decision.
2. The method of claim 1, wherein obtaining a predetermined decision period comprises, Acquiring vehicle information, wherein the vehicle information at least comprises communication delay, decision delay and data processing duration, When the sum of the communication delay, the decision delay and the data processing time length is less than the decision data acquisition period corresponding to the driving decision, determining that the decision multiple is a first character, When the sum of the communication delay, the decision delay and the data processing time length is determined to be greater than or equal to a decision data acquisition period corresponding to the driving decision, determining that the decision multiple is a second character, wherein the second character is greater than the first character, And for each driving decision, determining the product of the decision multiple and the decision data acquisition period corresponding to the driving decision as the decision period of the driving decision.
3. The method according to claim 1, wherein after acquiring the driving data determined in the decision data acquisition cycle corresponding to the driving decision that has ended, comprising: And updating the decision period and the decision space range of the space-time partition corresponding to the driving decision according to the driving data and the decision data acquisition period corresponding to the driving decision.
4. A method according to claim 3, wherein updating the decision period corresponding to the driving decision based on the driving data corresponding to the driving decision and the decision data acquisition period comprises: determining a first-level influence position and a second-level influence position in a decision space range corresponding to the driving decision, When the first-level influence position of the decision space range is determined to influence the vehicle according to the driving data, determining that the decision multiple is a third character, When the first-level influence position of the decision space range is determined to not influence the vehicle according to the driving data, and the second-level influence position of the decision space range is determined to influence the vehicle, determining a decision multiple as a fourth character, When the primary influence position and the secondary influence position of the decision space range are determined to have no influence on the vehicle according to the driving data, determining that the decision multiple is a first character or a second character, And determining the product of the decision multiple and the decision data acquisition period as an updated decision period.
5. A method according to claim 3, wherein updating the decision space range corresponding to the driving decision according to the driving data corresponding to the driving decision comprises: determining a driving speed according to driving data corresponding to the driving decision, Determining a braking distance corresponding to the running speed in a preset braking distance standard as a vehicle space length range, Updating the decision space range according to the vehicle space length range, Wherein the decision space range is composed of a plurality of vehicle space ranges, and each vehicle space range is the product of a vehicle space length range and a lane width range.
6. The method of claim 5, wherein determining a braking distance corresponding to the travel speed in a preset braking distance criterion as a vehicle space length range comprises: in response to a user operation, a security factor is determined, And determining the product of the safety coefficient and the braking distance corresponding to the running speed in a preset braking distance standard as a vehicle space length range.
7. The method of claim 1, wherein controlling the vehicle to travel in accordance with the driving instruction comprises: the running data of the vehicle is acquired, Determining a predicted travel path corresponding to the driving instruction, And when the existence of the object and the vehicle are at the same position at the same moment according to the running data and the predicted running path, re-planning and executing the driving instruction.
8. The method according to claim 1, wherein after controlling the running of the vehicle in accordance with the driving instruction, the method further comprises: the running data of the vehicle is acquired, Determining a space-time deviation value of a predicted running track corresponding to the driving instruction and a running track of the vehicle in the running process according to the running data and the driving instruction, wherein the space-time deviation value comprises a time deviation value and a space deviation value, When the space-time deviation value is determined to be larger than a preset space-time error threshold value, a driving instruction corresponding to the predicted driving track is re-planned according to the predicted driving track, And controlling the vehicle to run according to the re-planned driving instruction.
9. The method according to claim 1, wherein planning driving instructions corresponding to the driving decisions based on the decision driving data comprises: Respectively acquiring a destination level driving decision, a road level driving decision and a lane level driving decision which are being executed, wherein the driving decision comprises the destination level driving decision, the road level driving decision, the lane level driving decision and an action level driving decision, When planning the road-level driving decision, planning the road-level driving decision which does not conflict with the destination-level driving decision according to the decision driving data corresponding to the road-level driving decision and the destination-level driving decision, When planning the lane-level driving decision, planning the lane-level driving decision which does not conflict with the road-level driving decision according to the decision driving data corresponding to the lane-level driving decision and the road-level driving decision, And when planning the action-level driving decision, planning action-level driving decisions which do not conflict with the lane-level driving decision according to decision driving data corresponding to the action-level driving decision and the lane-level driving decision.
10. A vehicle control apparatus, characterized in that the apparatus comprises: an acquisition unit for acquiring a space-time partition corresponding to at least two preset driving decisions, wherein the space-time partition at least comprises a decision data acquisition period, a decision period and a decision space range, An acquisition unit for acquiring, for each driving decision, driving data determined in a decision data acquisition period corresponding to the driving decision that has ended when a decision period corresponding to the driving decision is reached, A determining unit configured to determine decision driving data corresponding to a decision space range of the driving decision from the driving data, A planning unit for planning driving instructions corresponding to the driving decisions according to the decision driving data, A control unit for controlling the vehicle to run according to the driving instruction; The device further comprises an acquisition subunit, a decision-making unit and a decision-making unit, wherein the acquisition subunit is used for determining at least one target vehicle sensor corresponding to each driving decision, respectively acquiring the data acquisition period of the at least one target vehicle sensor, and determining the maximum data acquisition period in the at least one data acquisition period as a decision-making data acquisition period; The acquisition subunit is used for acquiring sensor data acquired by the vehicle sensor in a decision data acquisition period corresponding to the driving decision which is ended and has the minimum time interval at the current moment, fitting the sensor data through a linear interpolation algorithm according to the sensor data acquired by the vehicle sensor in the decision data acquisition period corresponding to the driving decision which is ended and has the minimum time interval at the current moment, so as to obtain fitted sensor data corresponding to the current moment, and determining the obtained fitted sensor data as driving data; And the planning subunit is used for determining the distance to be driven of a part which is not driven in the predicted driving distance of the driving instruction according to the decision driving data and the driving instruction which is being executed when the driving decision is a lane-level driving decision, determining the driving instruction as a re-planned driving instruction when the distance to be driven is determined to be larger than a preset distance threshold value, and re-planning and executing the driving instruction according to the decision driving data corresponding to the lane-level driving decision when the distance to be driven is determined to be smaller than or equal to the distance threshold value.

Description

Vehicle control method, device, equipment and storage medium Technical Field The application belongs to the technical field of automobile electronics, and particularly relates to a vehicle control method, a device, equipment and a storage medium. Background With the development of automation, computer technology and electronics, some service providers have provided solutions for automatic driving of vehicles. In the process of automatically driving a vehicle, a vehicle control center generally controls the running of the vehicle based on data collected by sensors such as radar sensors, image sensors, and sound sensors. However, the data collected by the sensor is difficult to effectively utilize and has low efficiency. Disclosure of Invention The embodiment of the application provides a vehicle control method, a device and equipment, which can improve the utilization rate of data collected by a sensor. In one aspect, an embodiment of the present application provides a vehicle control method, including: Acquiring a space-time partition corresponding to at least two preset driving decisions respectively, wherein the space-time partition at least comprises a decision data acquisition period, a decision period and a decision space range, For each driving decision, when the decision period corresponding to the driving decision is reached, obtaining the driving data determined in the decision data acquisition period corresponding to the driving decision which is finished, Determining decision driving data corresponding to a decision space range of the driving decision from the driving data, Planning a driving instruction corresponding to the driving decision according to the decision driving data, And controlling the vehicle to run according to the driving instruction. Optionally, acquiring the decision data acquisition period includes: For each driving decision, determining at least one target vehicle sensor corresponding to the driving decision, Respectively acquiring data acquisition periods of the at least one target vehicle sensor, And determining the maximum data acquisition period in the at least one data acquisition period as a decision data acquisition period. Optionally, a predetermined decision period is obtained, including, Acquiring vehicle information, wherein the vehicle information at least comprises communication delay, decision delay and data processing duration, When the sum of the communication delay, the decision delay and the data processing time length is determined to be smaller than the unified data acquisition period, determining that the decision multiple is a first character, When the sum of the communication delay, the decision delay and the data processing time length is determined to be greater than or equal to the unified data acquisition period, determining that the decision multiple is a second character, wherein the second character is greater than the first character, And for each driving decision, determining the product of the decision multiple and the decision data acquisition period corresponding to the driving decision as the decision period of the driving decision. Optionally, obtaining driving data determined in a decision data acquisition period corresponding to the driving decision after the driving decision is finished includes: Sensor data acquired by a vehicle sensor during a decision data acquisition period corresponding to the driving decision that has ended and for which the time interval at the present moment is the smallest is acquired, Fitting the sensor data by a linear interpolation algorithm according to the sensor data acquired by the vehicle sensor in the unified data acquisition period to obtain fitted sensor data corresponding to the current moment, And determining the obtained fitting sensor data as driving data. Optionally, after obtaining the driving data determined in the decision data acquisition period corresponding to the driving decision after the driving decision is finished, the method includes: And updating the decision period and the decision space range of the space-time partition corresponding to the driving decision according to the driving data and the decision data acquisition period corresponding to the driving decision. Optionally, updating the decision period corresponding to the driving decision according to the driving data corresponding to the driving decision and the decision data acquisition period includes: determining a first-level influence position and a second-level influence position in a decision space range corresponding to the driving decision, When the first-level influence position of the decision space range is determined to influence the vehicle according to the driving data, determining that the decision multiple is a third character, When the first-level influence position of the decision space range is determined to not influence the vehicle according to the driving data, and the second-level influence position o