Search

CN-121999613-A - Method for determining avoidance tag, electronic device, apparatus and storage medium

CN121999613ACN 121999613 ACN121999613 ACN 121999613ACN-121999613-A

Abstract

The electronic equipment comprises a processor, first state information of a vehicle at the time of image data acquisition and second state information of an obstacle in the image data at the time of acquisition are determined, multi-frame target images are determined based on the first state information, the second state information and the multi-frame image data, and avoidance labels of the obstacle are determined based on the first state information, the second state information and the multi-frame target images and are used for indicating avoidance operation of the vehicle for the obstacle. According to the technical scheme, the multi-frame target image is screened out from the multi-frame image data, and then the avoidance labels of the obstacles in the multi-frame target image are determined based on the state information of the vehicle and the obstacles at the acquisition time of the multi-frame target image, so that automatic labeling of the avoidance labels can be accurately and simply realized by using posterior information through a mode of first screening and then accurate labeling.

Inventors

  • FAN KUN
  • TAI LEI
  • MAO WEIBO

Assignees

  • 上海安亭地平线智能交通技术有限公司

Dates

Publication Date
20260508
Application Date
20260209

Claims (20)

  1. 1. An electronic device comprising at least one processor configured to: Determining multi-frame image data; determining first state information of a vehicle at the image data acquisition time and second state information of an obstacle in the image data at the acquisition time; Determining a multi-frame target image based on the first state information, the second state information and the multi-frame image data; and determining an avoidance tag of the obstacle based on the first state information, the second state information and the multi-frame target image, wherein the avoidance tag is used for indicating the avoidance operation of the self-vehicle on the obstacle.
  2. 2. The electronic device of claim 1, wherein the processor performs determining a multi-frame target image based on the first state information, the second state information, and the multi-frame image data, specifically configured to: Determining a target frame interval based on the first state information and the second state information, wherein the target frame interval is a frame interval in which interaction possibility exists between the vehicle and the obstacle; and determining the multi-frame target image based on the target frame interval and the multi-frame image data.
  3. 3. The electronic device of claim 2, wherein the processor performs determining a target frame interval based on the first state information and the second state information, specifically configured to: Determining an interaction key frame from the multi-frame image data based on the first state information and the second state information; determining index information corresponding to the interaction key frame based on the interaction key frame; and determining the target frame interval based on the index information.
  4. 4. The electronic device of claim 3, wherein the processor performs determining an interaction keyframe from the multi-frame image data based on the first status information and the second status information, specifically configured to: Determining a first image from the multi-frame image data based on the first state information and the second state information, wherein the longitudinal coordinates of the vehicle and the obstacle in a preset coordinate system have the same value at the acquisition time of the first image; determining a first speed of the own vehicle and a second speed of the obstacle based on the first state information, the second state information, and the first image; and determining the first image as the interaction key frame in response to the first speed being greater than or equal to the second speed.
  5. 5. The electronic device of claim 4, wherein the processor performs determining an interaction key frame from the multi-frame image data based on the first status information and the second status information, and is further specifically configured to: determining a second image from the multi-frame image data based on the first state information and the second state information in response to the fact that the interaction key frame is not determined, wherein the obstacle is the nearest vehicle in the same direction of the lane where the own vehicle is located and the own vehicle and the obstacle are in a static state at the same time at the time of collecting the second image; and determining the second image as the interaction key frame.
  6. 6. The electronic device of claim 5, wherein the processor performs determining an interaction key frame from the multi-frame image data based on the first status information and the second status information, and is further specifically configured to: in response to the interaction key frame not being determined and the first speed being less than the second speed, the first image is determined to be the interaction key frame.
  7. 7. The electronic device of claim 5, wherein the processor performs determining an interaction key frame from the multi-frame image data based on the first status information and the second status information, and is further specifically configured to: Determining a plurality of frames of third images from the plurality of frames of image data based on the first state information and the second state information in response to the fact that the interaction key frame is not determined, wherein the transverse coordinates of the vehicle and the obstacle in the preset coordinate system are the same in value at the time of collecting the third images; and determining a fourth image based on the multi-frame third image, and determining the fourth image as the interaction key frame.
  8. 8. The electronic device of claim 5, wherein the processor performs determining an interaction key frame from the multi-frame image data based on the first status information and the second status information, and is further specifically configured to: Determining the longitudinal distance between the vehicle and the obstacle in the preset coordinate system at the acquisition time of the multi-frame image data in response to the fact that the interaction key frame is not determined and the longitudinal coordinate of the obstacle in the preset coordinate system at the acquisition time of the multi-frame image data is larger than the longitudinal coordinate of the vehicle; And determining a target longitudinal distance based on a plurality of longitudinal distances, and determining an image corresponding to the target longitudinal distance as the interaction key frame.
  9. 9. The electronic device of claim 1, wherein the processor performs determining an avoidance tag of the obstacle based on the first state information, the second state information, and the multi-frame target image, specifically configured to: Determining a lateral distance between the vehicle and the obstacle in a preset coordinate system at the acquisition time of the multi-frame target image and a relative position between the vehicle and the obstacle based on the first state information and the second state information; And determining an avoidance tag of the obstacle in the multi-frame target image as a first avoidance tag in response to the transverse distance meeting a first distance relation or the relative position meeting a position relation.
  10. 10. The electronic device of claim 9, wherein the processor performs determining the avoidance tag of the obstacle based on the first status information, the second status information, and the multi-frame target image, and is further specifically configured to: And responding to the transverse distance meeting a second distance relation, and determining that the avoidance label of the obstacle in the multi-frame target image is a second avoidance label.
  11. 11. The electronic device of claim 10, the processor further configured to: Determining a longitudinal distance between the vehicle and the obstacle in the preset coordinate system at the acquisition time of the multi-frame target image, and a third speed of the vehicle and a fourth speed of the obstacle based on the first state information and the second state information; determining a category attribute of the obstacle; And determining that the avoidance tag of the obstacle in the multi-frame target image is a second sub-avoidance tag based on the category attribute, the transverse distance, the longitudinal distance, the third speed and the fourth speed.
  12. 12. The electronic device of claim 10, wherein the processor performs determining the avoidance tag of the obstacle based on the first status information, the second status information, and the multi-frame target image, and is further specifically configured to: And determining that the avoidance tag of the obstacle in the multi-frame target image is a third avoidance tag in response to the transverse distance does not meet the first distance relationship and the second distance relationship and the relative position does not meet the position relationship.
  13. 13. The electronic device of claim 12, the processor further configured to: Determining a first moment and a second moment based on the first state information and the second state information, wherein the first moment is the moment when the longitudinal coordinates of the vehicle and the obstacle in the preset coordinate system have the same value, and the second moment is the starting time of the transverse movement of the vehicle; And determining that an avoidance tag of the obstacle in the multi-frame target image is a third sub-avoidance tag based on the first moment, the second moment, the transverse distance, the longitudinal distance and the relative position.
  14. 14. A method of determining an avoidance tag, the method comprising: Determining multi-frame image data; determining first state information of a vehicle at the image data acquisition time and second state information of an obstacle in the image data at the acquisition time; Determining a multi-frame target image based on the first state information, the second state information and the multi-frame image data; and determining an avoidance tag of the obstacle based on the first state information, the second state information and the multi-frame target image, wherein the avoidance tag is used for indicating the avoidance operation of the self-vehicle on the obstacle.
  15. 15. The method of claim 14, wherein the determining a multi-frame target image based on the first state information, the second state information, and the multi-frame image data comprises: Determining a target frame interval based on the first state information and the second state information, wherein the target frame interval is a frame interval in which interaction possibility exists between the vehicle and the obstacle; and determining the multi-frame target image based on the target frame interval and the multi-frame image data.
  16. 16. The method of claim 15, wherein the determining a target frame interval based on the first status information and the second status information comprises: Determining an interaction key frame from the multi-frame image data based on the first state information and the second state information; determining index information corresponding to the interaction key frame based on the interaction key frame; and determining the target frame interval based on the index information.
  17. 17. The method of claim 14, wherein the determining the avoidance tag of the obstacle based on the first state information, the second state information, and the multi-frame target image comprises: Determining a lateral distance between the vehicle and the obstacle in a preset coordinate system at the acquisition time of the multi-frame target image and a relative position between the vehicle and the obstacle based on the first state information and the second state information; And determining an avoidance tag of the obstacle in the multi-frame target image as a first avoidance tag in response to the transverse distance meeting a first distance relation or the relative position meeting a position relation.
  18. 18. The method of claim 17, wherein the determining the avoidance tag of the obstacle based on the first state information, the second state information, and the multi-frame target image further comprises: And responding to the transverse distance meeting a second distance relation, and determining that the avoidance label of the obstacle in the multi-frame target image is a second avoidance label.
  19. 19. A determination device for avoiding a tag, comprising: The data determining module is used for determining multi-frame image data; the state determining module is used for determining first state information of the own vehicle at the image data acquisition time and second state information of the obstacle in the image data at the acquisition time; an image determining module, configured to determine a multi-frame target image based on the first state information, the second state information, and the multi-frame image data; The tag determining module is used for determining an avoidance tag of the obstacle based on the first state information, the second state information and the multi-frame target image, wherein the avoidance tag is used for indicating the avoidance operation of the vehicle on the obstacle.
  20. 20. A computer readable storage medium storing a computer program for performing the method of determining an avoidance tag of any of the preceding claims 14-18.

Description

Method for determining avoidance tag, electronic device, apparatus and storage medium Technical Field The disclosure relates to the technical field of intelligent driving, and in particular relates to a method for determining avoidance labels, electronic equipment, a device and a storage medium. Background The development of the intelligent driving related technology is more and more mature, and the vehicle needs to accurately sense the surrounding environment in the driving process so as to ensure the driving safety. However, in a complex traffic scene, how to accurately judge the interaction relationship between the vehicle and the obstacle and formulate a reasonable avoidance strategy according to the interaction relationship is one of the great challenges faced by the current technology. Among other things, avoidance strategies include bypassing an obstacle or following an obstacle, and if bypassing an obstacle, bypassing from the left or right, etc. At present, an intelligent driving system generally adopts an end-to-end model, and the model can directly output instructions required for realizing avoidance and vehicle planning control according to environmental information. However, when the end-to-end model is trained, the related technology generally relies on a manual labeling mode to obtain the needed avoidance labels, and the mode is time-consuming and labor-consuming, is easily affected by subjective factors, and therefore consistency and accuracy of labeling results are difficult to guarantee. Disclosure of Invention In order to solve the technical problems, the present disclosure provides a method for determining an avoidance tag, an electronic device, an apparatus, and a storage medium. In a first aspect of the disclosure, an embodiment provides an electronic device including at least one processor configured to determine multi-frame image data, determine first state information of a vehicle at an image data acquisition time and second state information of an obstacle in the image data at the acquisition time, determine multi-frame target images based on the first state information, the second state information and the multi-frame image data, and determine avoidance labels of the obstacle based on the first state information, the second state information and the multi-frame target images, wherein the avoidance labels are used for indicating avoidance operations of the vehicle for the obstacle. An embodiment of a second aspect of the present disclosure provides a method for determining an avoidance tag, including determining multi-frame image data, determining first state information of a host vehicle at an image data acquisition time and second state information of an obstacle in the image data at the acquisition time, determining multi-frame target images based on the first state information, the second state information and the multi-frame image data, and determining an avoidance tag of the obstacle based on the first state information, the second state information and the multi-frame target images, wherein the avoidance tag is used for indicating avoidance operation of the host vehicle for the obstacle. The embodiment of the third aspect of the disclosure provides a device for determining an avoidance tag, which comprises a data determining module, a state determining module, an image determining module and a tag determining module, wherein the data determining module is used for determining multi-frame image data, the state determining module is used for determining first state information of a vehicle at the time of image data acquisition and second state information of an obstacle in the image data at the time of acquisition, the image determining module is used for determining multi-frame target images based on the first state information, the second state information and the multi-frame image data, and the tag determining module is used for determining the avoidance tag of the obstacle based on the first state information, the second state information and the multi-frame target images, and the avoidance tag is used for indicating avoidance operation of the vehicle for the obstacle. In a fourth aspect of the present disclosure, there is provided a computer readable storage medium storing a computer program for executing the method for determining an avoidance tag provided in the second aspect. In a fifth aspect of the present disclosure, an embodiment of the present disclosure provides an electronic device, where the electronic device includes a processor, a memory configured to store executable instructions of the processor, and a processor configured to read the executable instructions from the memory and execute the instructions to implement the method for determining an avoidance tag provided in the second aspect of the present disclosure. A sixth aspect of the present disclosure provides a computer program product, which when executed by a processor, performs the method for determi