Search

CN-121999461-A - Target detection method and device, whole vehicle controller and vehicle

CN121999461ACN 121999461 ACN121999461 ACN 121999461ACN-121999461-A

Abstract

The embodiment of the application provides a target detection method, a target detection device, a whole vehicle controller and a vehicle, wherein first target detection result sets corresponding to first fusion targets at a plurality of moments are obtained, the first fusion targets are obtained through fusion based on all first target detection results in the first target detection result sets, target associated cost values of a sensor set formed by sensors corresponding to the first target detection results are determined based on the first target detection result sets, the target sensors are determined from the sensor set based on the target associated cost values corresponding to the sensor set at the plurality of moments, and the first targets are independently determined through the first target detection results of the target sensors. The method can detect the target fusion result corresponding to the sensor for generating the fusion target, and timely find out the sensor which is not suitable for continuously generating the fusion target, so that the target detection results of the sensors are stripped to independently detect the target, and the accuracy of target detection is improved.

Inventors

  • HAN RUTAO

Assignees

  • 毫末智行科技有限公司

Dates

Publication Date
20260508
Application Date
20241105

Claims (10)

  1. 1. A method of target detection, the method comprising: Acquiring first target detection result sets corresponding to first fusion targets at a plurality of moments respectively, wherein the first fusion targets are obtained by fusion based on all first target detection results in the first target detection result sets; Determining target associated cost values of a sensor set formed by sensors respectively corresponding to the first target detection results based on the first target detection result set; Determining a target sensor from the sensor set based on target associated cost values respectively corresponding to the sensor set at the plurality of moments; and independently determining a first target through a first target detection result of the target sensor.
  2. 2. The method according to claim 1, wherein determining the target associated cost value of the sensor set formed by the sensors respectively corresponding to the first target detection results based on the first target detection result set includes: determining a first association cost value between a first sensor and a second sensor outside the first sensor in the sensor set based on the first target detection result; and determining the target associated cost value corresponding to the sensor set based on the first associated cost value and the number of sensors in the sensor set.
  3. 3. The method of claim 1, wherein the determining the target sensor from the set of sensors based on the target associated cost values respectively corresponding to the set of sensors at the plurality of times comprises: determining a change trend of the association cost based on the target association cost values respectively corresponding to the sensor set at the plurality of moments; And under the condition that the association cost change trend is greater than or equal to a first threshold value, determining each sensor in the sensor set as a target sensor.
  4. 4. The method of claim 1, wherein the target association cost value comprises a first association cost value between a first sensor and a second sensor other than the first sensor in the set of sensors, the determining a target sensor from the set of sensors based on the target association cost values respectively corresponding to the set of sensors at the plurality of times, comprising: Determining a first number corresponding to a first association cost value of a first sensor less than or equal to a second threshold; determining the first sensor as the target sensor when the first number is greater than or equal to a second number, wherein the second number is obtained by subtracting a preset number from the number of sensors in the sensor set; and reconstructing a sensor set based on the first sensors under the condition that the first number is smaller than or equal to the third number, and fusing first target detection results corresponding to each sensor in the reconstructed sensor set to update a first fusion target, wherein the third number is smaller than the second number.
  5. 5. The method of claim 2, wherein the determining a first associated cost value between a first sensor and a second sensor, other than the first sensor, in the set of sensors based on the first target detection result comprises: Determining a target mahalanobis distance and a target state similarity based on a first target detection result of the first sensor and a first target detection result of the second sensor; and carrying out weighted summation on the target mahalanobis distance and the target state similarity to obtain the first association cost value.
  6. 6. The method of claim 2, wherein the determining a first associated cost value between a first sensor and a second sensor, other than the first sensor, in the set of sensors based on the first target detection result comprises: determining a first target speed parameter based on a first target detection result of the first sensor; determining a second target speed parameter based on a first target detection result of the second sensor; And determining the first associated cost value according to the difference between the first target speed parameter and the second target speed parameter.
  7. 7. The method according to claim 1, wherein the method further comprises: Acquiring a second fusion target within a preset range of the first fusion target under the condition that the first fusion target belongs to a preset target type and the target distance of the first fusion target is smaller than the preset distance; Determining a second association cost value between a third sensor of a first sensor type in the sensor set and a fourth sensor of a second sensor type in the sensor set according to the first target detection result; determining a third association cost value between the third sensor and the fourth sensor according to the first target detection result and a second target detection result corresponding to the second fusion target; And under the condition that the third association cost is smaller than the second association cost, establishing a fusion association relation between a first target detection result corresponding to the third sensor and a second target detection result corresponding to the fourth sensor so as to fuse the first target detection result corresponding to the third sensor and the second target detection result corresponding to the fourth sensor to obtain a third fusion target.
  8. 8. An object detection device, the device comprising: The acquisition module is used for acquiring first target detection result sets respectively corresponding to first fusion targets at a plurality of moments, wherein the first fusion targets are obtained by fusion based on all first target detection results in the first target detection result sets; The cost value module is used for determining the target associated cost value of a sensor set formed by the sensors respectively corresponding to the first target detection results based on the first target detection result set; The sensor module is used for determining a target sensor from the sensor set based on the target associated cost values respectively corresponding to the sensor set at the plurality of moments; And the detection module is used for independently determining a first target according to a first target detection result of the target sensor.
  9. 9. A vehicle control unit, characterized in that it comprises a memory, a processor and a computer program stored on the memory and executable on the processor, which processor implements the object detection method according to any one of claims 1 to 7 when executing the computer program.
  10. 10. A vehicle comprising a vehicle control unit according to claim 9.

Description

Target detection method and device, whole vehicle controller and vehicle Technical Field The present application relates to the field of target detection technologies, and in particular, to a target detection method and apparatus, a vehicle control unit, and a vehicle. Background With the continuous development of technology, automatic driving and assisted driving technologies are increasingly applied to modern vehicles. In order to improve the recognition and navigation capabilities of vehicles in complex environments, fusion target detection techniques are receiving increasing attention. In the related art, generally, ID matching or the like is adopted to match target detection results corresponding to the same entity from each sensor, and then the matched target detection results are fused to determine a fusion target. However, in the continuous operation process of the sensor, the corresponding relation between the target detection result and the real object, the accuracy of the target detection result and the like are changed, so that the accuracy of the fusion target obtained by fusion based on the matching relation is poor. Disclosure of Invention In view of the above, the present application aims to provide a target detection method, a device, a whole vehicle controller and a vehicle, so as to solve the problem of poor target detection accuracy of multi-sensor fusion in the prior art. In order to achieve the above purpose, the technical scheme of the application is realized as follows: in a first aspect, the present application provides a target detection method, the method comprising: Acquiring first target detection result sets corresponding to first fusion targets at a plurality of moments respectively, wherein the first fusion targets are obtained by fusion based on all first target detection results in the first target detection result sets; Determining target associated cost values of a sensor set formed by sensors respectively corresponding to the first target detection results based on the first target detection result set; Determining a target sensor from the sensor set based on target associated cost values respectively corresponding to the sensor set at the plurality of moments; and independently determining a first target through a first target detection result of the target sensor. Optionally, the determining, based on the first target detection result set, the target associated cost value of the sensor set formed by the sensors corresponding to the first target detection results respectively includes: determining a first association cost value between a first sensor and a second sensor outside the first sensor in the sensor set based on the first target detection result; and determining the target associated cost value corresponding to the sensor set based on the first associated cost value and the number of sensors in the sensor set. Optionally, the determining, based on the target associated cost values of the sensor set corresponding to the multiple moments respectively, the target sensor from the sensor set includes: determining a change trend of the association cost based on the target association cost values respectively corresponding to the sensor set at the plurality of moments; And under the condition that the association cost change trend is greater than or equal to a first threshold value, determining each sensor in the sensor set as a target sensor. Optionally, the target association cost value includes a first association cost value between a first sensor and a second sensor other than the first sensor in the sensor set, and determining, based on the target association cost values of the sensor set corresponding to the multiple moments respectively, a target sensor from the sensor set includes: Determining a first number corresponding to a first association cost value of a first sensor less than or equal to a second threshold; determining the first sensor as the target sensor when the first number is greater than or equal to a second number, wherein the second number is obtained by subtracting a preset number from the number of sensors in the sensor set; and reconstructing a sensor set based on the first sensors under the condition that the first number is smaller than or equal to the third number, and fusing first target detection results corresponding to each sensor in the reconstructed sensor set to update a first fusion target, wherein the third number is smaller than the second number. Optionally, the determining, based on the first target detection result, a first association cost value between a first sensor and a second sensor other than the first sensor in the sensor set includes: Determining a target mahalanobis distance and a target state similarity based on a first target detection result of the first sensor and a first target detection result of the second sensor; and carrying out weighted summation on the target mahalanobis distance and the target state simil