Search

CN-121973766-A - Vehicle perception boundary self-adaptive adjusting method, device, product and medium

CN121973766ACN 121973766 ACN121973766 ACN 121973766ACN-121973766-A

Abstract

A vehicle perception boundary self-adaptive adjusting method, device, product and medium relate to the technical field of intelligent driving. The method comprises the steps of obtaining kinematic state data, environmental weather data and detection data of a vehicle, identifying an obstacle according to the detection data, determining the position of the obstacle relative to a space area of the vehicle body, determining a corresponding area sensitivity coefficient, determining a source sensor for detecting the obstacle from a plurality of sensors, determining a sensor sensitivity coefficient, determining a group of environmental correction factors according to the environmental weather data, calculating a dynamic boundary expansion amount according to the kinematic state data, carrying out fusion calculation based on the area sensitivity coefficient, the sensor sensitivity coefficient, the environmental correction factors and the dynamic boundary expansion amount, generating a dynamic perception boundary for the obstacle, and generating a vehicle safety response instruction when the obstacle is determined to be located in the dynamic perception boundary. Has the effect of improving the safety of the vehicle in complex environments.

Inventors

  • ZHU XING
  • JIANG YINGHAO

Assignees

  • 武汉未来幻影科技有限公司

Dates

Publication Date
20260505
Application Date
20260202

Claims (10)

  1. 1. A method for adaptively adjusting a perceived boundary of a vehicle, the method comprising: acquiring kinematic state data, environmental weather data, and detection data of a plurality of sensors arranged around a vehicle body; Identifying obstacles around the vehicle body according to the detection data, determining the position of the obstacles relative to the spatial region of the vehicle body, and determining a region sensitivity coefficient corresponding to the position of the spatial region according to a preset spatial sensitivity mapping relation; determining a source sensor detecting an obstacle from a plurality of sensors, and determining a sensor sensitivity coefficient according to a state configuration of the source sensor; Determining a group of environment correction factors according to the environmental meteorological data, and calculating dynamic boundary expansion according to the kinematic state data; Performing fusion calculation based on the regional sensitivity coefficient, the sensor sensitivity coefficient, the environment correction factor and the dynamic boundary expansion amount to generate a dynamic perception boundary aiming at the obstacle; When it is determined that the obstacle is within the dynamic perception boundary, a vehicle safety response instruction is generated.
  2. 2. The vehicle perception boundary adaptive adjustment method according to claim 1, wherein the determining a source sensor that detects an obstacle from a plurality of sensors, and determining a sensor sensitivity coefficient according to a state configuration of the source sensor, specifically includes: constructing a sensor independent sensitivity vector comprising a plurality of dimensions, wherein each dimension of the independent sensitivity vector corresponds to a sensitivity weight parameter of a sensor; dividing the plurality of sensors into at least one logical sensor group according to the physical installation position; determining an operational health status of each source sensor based on the status configuration; If the operation health state is a normal state, setting the sensitivity weight parameter of the corresponding source sensor in the independent sensitivity vector as a standard value; If the operation health state is an occlusion state, setting a sensitivity weight parameter of a corresponding source sensor in the independent sensitivity vector to be a preset attenuation value; If the operation health state is a functional failure state, setting a sensitivity weight parameter of a corresponding source sensor in the independent sensitivity vector as an invalid identification value; and acquiring sensitivity weight parameters of all sensors in the logical sensor group to which the source sensor belongs, removing parameters of invalid identification values, and performing aggregate calculation on the rest sensitivity weight parameters to obtain a sensor sensitivity coefficient.
  3. 3. The method for adaptively adjusting a perceived boundary of a vehicle according to claim 1, wherein said determining a set of environmental correction factors based on environmental weather data, in particular, comprises: Performing characteristic analysis on the environmental meteorological data to determine the interference level of the current environment; Retrieving a sensitivity adjustment strategy matched with the interference level from a preset environment strategy library; According to the sensitivity adjustment strategy, weight correction values corresponding to different types of sensors in the plurality of sensors are determined, and each weight correction value forms a set of environment correction factors.
  4. 4. The vehicle-perceived-boundary adaptive adjustment method of claim 1, wherein the calculating the dynamic boundary expansion from the kinematic state data specifically includes: extracting the running speed, steering angle and current gear information of the vehicle from the kinematic state data; Calculating basic expansion amplitude based on the running speed and the steering angle, wherein the basic expansion amplitude is positively correlated with the running speed and the steering angle; Determining a boundary expansion direction according to the current gear information: If the current gear is a forward gear, setting the boundary expansion direction as the same steering side indicated by the steering angle based on a preset inner wheel difference protection principle; If the current gear is a reverse gear, setting the boundary expansion direction as a steering opposite side indicated by a steering angle based on a preset headstock outward swing compensation principle so as to cover a headstock outward swing area; and combining the basic expansion amplitude and the boundary expansion direction to generate the dynamic boundary expansion quantity.
  5. 5. The method for adaptively adjusting a vehicle sensing boundary according to claim 4, wherein the generating a dynamic sensing boundary for an obstacle based on the fusion calculation of the regional sensitivity coefficient, the sensor sensitivity coefficient, the environment correction factor and the dynamic boundary expansion amount comprises: Acquiring a preset vehicle reference safety boundary; Scaling the part corresponding to the space region position in the reference safety boundary by using the region sensitivity coefficient to obtain a first intermediate boundary; Weighting correction is carried out on the first intermediate boundary by using a sensor sensitivity coefficient and an environment correction factor to obtain a second intermediate boundary; And the dynamic boundary expansion amount is added to the direction of the second middle boundary corresponding to the boundary expansion direction, so that the dynamic perception boundary is obtained.
  6. 6. The method for adaptively adjusting a vehicle perception boundary according to claim 1, wherein when it is determined that an obstacle is located within a dynamic perception boundary, a vehicle safety response command is generated, specifically comprising: calculating a real-time Euclidean distance between an obstacle and a preset vehicle contour, and acquiring a boundary threshold value of a dynamic perception boundary in the direction of the obstacle; when the real-time Euclidean distance is smaller than the boundary threshold value, determining that the obstacle is positioned in the dynamic perception boundary, and calculating the difference value between the boundary threshold value and the real-time Euclidean distance as the boundary invasion depth of the obstacle; And generating a vehicle safety response instruction matched with the risk level based on the boundary invasion depth, wherein the vehicle safety response instruction is an alarm prompt instruction, a deceleration control instruction or an emergency braking instruction.
  7. 7. The method for adaptively adjusting the perceived boundary of a vehicle according to claim 6, wherein the generating of the vehicle safety response command matched with the risk level based on the depth of intrusion of the boundary specifically comprises: Acquiring a preset first risk threshold and a preset second risk threshold, wherein the second risk threshold is larger than the first risk threshold; If the boundary invasion depth is smaller than or equal to the first risk threshold, judging that the risk level is a low risk level, and generating an alarm prompt instruction; If the boundary intrusion depth is larger than the first risk threshold and smaller than or equal to the second risk threshold, judging that the risk level is a middle risk level, and generating a deceleration control instruction; and if the boundary invasion depth is larger than the second risk threshold value, judging that the risk level is a high risk level, and generating an emergency braking instruction.
  8. 8. An electronic device for a vehicle-aware boundary adaptive adjustment method, characterized in that the electronic device comprises one or more processors and a memory, the memory being coupled to the one or more processors, the memory being for storing computer program code comprising computer instructions, the one or more processors invoking the computer instructions to cause the electronic device to perform the method according to any of claims 1-7.
  9. 9. A computer program product comprising instructions which, when run on an electronic device of a method of adaptive adjustment of vehicle perceived boundaries, cause the electronic device to perform the method of any of claims 1-7.
  10. 10. A computer readable storage medium comprising instructions which, when run on an electronic device of a vehicle-aware boundary adaptive adjustment method, cause the electronic device to perform the method of any of claims 1-7.

Description

Vehicle perception boundary self-adaptive adjusting method, device, product and medium Technical Field The application relates to the technical field of intelligent driving, in particular to a vehicle perception boundary self-adaptive adjusting method, device, product and medium. Background With the rapid development of artificial intelligence and automatic driving technology, intelligent driving systems have gradually moved from laboratory research to practical application, becoming an important development direction in the future traffic field. The intelligent driving vehicle senses the surrounding environment through various sensors (such as a camera, a laser radar, a millimeter wave radar and the like) so as to realize the identification and avoidance of the obstacle, thereby ensuring the driving safety. In this process, the accurate setting of the vehicle perceived boundary is directly related to the safety and reliability of the intelligent driving system. At present, in the field of intelligent driving vehicle perception boundary test, a Chinese patent with publication number CN118746445B discloses intelligent driving vehicle perception boundary test equipment and a method thereof, according to the method, the perceived boundary performance of the vehicle in different scenes is tested by adjusting the height and the light intensity of the testing equipment and simulating different weather environments (such as spraying simulation rainy days), so that the perceived capability of the intelligent driving system is estimated. However, during actual road running, the environment where the vehicle is located dynamically changes, and different weather conditions (such as rain and fog will reduce the detection distance of the sensor), different vehicle movement states (such as a greater safety distance required for high-speed running), and different spatial regions (such as different hazard levels in front of and at the side of the vehicle) will all affect the reasonable range of the perceived boundary. Therefore, the perceived boundary of the vehicle is evaluated only by means of a static test scene, the requirement of dynamic self-adaptive adjustment of the perceived boundary in an actual road environment is difficult to meet, and the safety of the vehicle in a complex environment is easy to be lowered. Disclosure of Invention The application provides a vehicle perception boundary self-adaptive adjusting method, equipment, a product and a medium, which have the effect of improving the safety of a vehicle in a complex environment. In a first aspect of the present application, a method for adaptively adjusting a perceived boundary of a vehicle is provided, which specifically includes: acquiring kinematic state data, environmental weather data, and detection data of a plurality of sensors arranged around a vehicle body; Identifying obstacles around the vehicle body according to the detection data, determining the position of the obstacles relative to the spatial region of the vehicle body, and determining a region sensitivity coefficient corresponding to the position of the spatial region according to a preset spatial sensitivity mapping relation; determining a source sensor detecting an obstacle from a plurality of sensors, and determining a sensor sensitivity coefficient according to a state configuration of the source sensor; Determining a group of environment correction factors according to the environmental meteorological data, and calculating dynamic boundary expansion according to the kinematic state data; Performing fusion calculation based on the regional sensitivity coefficient, the sensor sensitivity coefficient, the environment correction factor and the dynamic boundary expansion amount to generate a dynamic perception boundary aiming at the obstacle; When it is determined that the obstacle is within the dynamic perception boundary, a vehicle safety response instruction is generated. According to the technical scheme, kinematic state data, environmental meteorological data and detection data of a plurality of sensors arranged around a vehicle body are firstly obtained, a comprehensive data basis is provided for subsequent perception boundary adjustment, then obstacles around the vehicle body are identified according to the detection data, the positions of the obstacles in the spatial region relative to the vehicle body are determined, regional sensitivity coefficients corresponding to the positions of the spatial region are determined according to a preset spatial sensitivity mapping relation, so that the obstacles in different spatial positions can obtain sensitivity weights matched with the dangerous degrees of the obstacles, further, a source sensor for detecting the obstacles is determined from the plurality of sensors, the sensor sensitivity coefficients are determined according to the state configuration of the source sensor, so that the perception boundary can be specifically adjusted accor