CN-116106866-B - Mining vehicle-mounted perception system performance degradation recognition system and method

CN116106866BCN 116106866 BCN116106866 BCN 116106866BCN-116106866-B

Abstract

The invention discloses a mining vehicle-mounted sensing system performance degradation identification system and method, comprising an environment sample feature memory storage module, an environment feature acquisition and identification module, a performance judgment module and an alarm module, wherein the environment sample feature memory storage module is used for carrying out pre-environment feature acquisition and processing by a vehicle-mounted sensing system at a marked position in the environment and storing a result as a priori sample feature, the environment feature acquisition and identification module is used for acquiring operation data in the vehicle operation process, carrying out vehicle-mounted sensing data acquisition when the vehicle runs to the marked position and converting the vehicle-mounted sensing data into operation process sample features, the performance judgment module is used for comparing the priori sample features with the operation process sample features to obtain differences of the priori sample features and the operation process sample features, judging that the result is abnormal when the differences exceed a threshold value, otherwise, judging that the result is normal, and the alarm module is used for sending an early warning signal when the judging that the result is abnormal. Early warning is timely carried out when the performance of the laser radar is reduced, and guarantee is provided for the safe operation of unmanned mine cards.

Inventors

TANG JIANLIN
FU QIANG
REN LIANGCAI

Assignees

江苏徐工工程机械研究院有限公司
江苏徐工国重实验室科技有限公司

Dates

Publication Date: 20260505
Application Date: 20221122

Claims (8)

1. The mining vehicle-mounted perception system performance degradation recognition system is characterized by comprising an environment sample feature memory storage module, an environment feature acquisition and recognition module, a performance judgment module and an alarm module; the environment sample feature memory storage module is used for carrying out pre-environment feature acquisition and processing on the marked positions in the environment by utilizing the vehicle-mounted sensing system and storing the results as priori sample features; The environment characteristic acquisition and identification module is used for acquiring operation data in the operation process of the vehicle, acquiring vehicle-mounted sensing data when the vehicle runs to the marked position, and converting the vehicle-mounted sensing data into operation process sample characteristics; the performance judging module is used for receiving the prior sample characteristics from the environment sample characteristic memory storage module and the operation process sample characteristics from the environment characteristic acquisition and identification module, comparing the prior sample characteristics with the operation process sample characteristics to obtain differences between the prior sample characteristics and the operation process sample characteristics, judging that the result is abnormal when the differences exceed a threshold value, and judging that the result is normal otherwise; The alarm module is used for receiving the judging result of the performance judging module and sending out an early warning signal when the judging result is abnormal; the environmental characteristic acquisition and recognition module comprises a position judgment sub-module, an ROI region filtering sub-module, a ground segmentation sub-module and a recognition sub-module; The position judging sub-module is used for receiving the position information of the mine card sent by the mine card positioning system, comparing the position information with the position information of the marked position, and sending execution information to the ROI area filtering sub-module when the two pieces of information are consistent; The ROI region filtering sub-module is used for receiving the execution information sent by the position judging module, reading in the point cloud data of the radar, only preserving the point cloud in the ROI region according to the coordinate value method of limiting the point cloud data, and sending the point cloud to the ground segmentation sub-module; The ground segmentation submodule is used for receiving the point cloud data sent by the ROI region filtering submodule, obtaining ground point cloud and obstacle point cloud data by using a RANSAC method, and sending the obstacle point cloud data to the identification submodule; The identification sub-module is used for receiving the point cloud data sent by the ground segmentation sub-module, calculating the number of the point clouds corresponding to the obstacle, obtaining the sample characteristics of the operation process, and sending the sample characteristics of the operation process to the performance judging module.
2. The mining vehicle-mounted sensing system performance degradation identification system according to claim 1 is characterized in that the prior sample is characterized by the number of point clouds of obstacles in the environment, the vehicle-mounted sensing system comprises a laser radar, the laser radar is used for acquiring environment data, and the number of the point clouds of the obstacles is obtained through processing.
3. The mining vehicle-mounted perception system performance degradation identification system of claim 1, wherein the operational process sample is characterized by a number of point clouds of obstacles in the environment.
4. The mining vehicle-mounted sensing system performance degradation identification system of claim 1, wherein the difference between the a priori sample characteristics and the run-time sample characteristics is an absolute value of the difference.
5. A mining vehicle-mounted perception system performance degradation recognition method is characterized by comprising the following steps of: the environment sample feature memory storage module is used for carrying out pre-environment feature collection and processing on the marked positions in the environment by using a vehicle-mounted sensing system, and storing the results as priori sample features; the environment characteristic acquisition and identification module acquires operation data in the operation process of the vehicle, and when the vehicle runs to the marked position, the vehicle-mounted sensing data acquisition is carried out and converted into operation process sample characteristics; the performance judging module receives the prior sample characteristics from the environment sample characteristic memory storage module and the operation process sample characteristics from the environment characteristic acquisition and identification module, compares the prior sample characteristics with the operation process sample characteristics to obtain differences between the prior sample characteristics and the operation process sample characteristics, judges that the result is abnormal when the differences exceed a threshold value, and judges that the result is normal; The alarm module receives the judging result of the performance judging module, and sends out an early warning signal when the judging result is abnormal; the environmental characteristic acquisition and recognition module comprises a position judgment sub-module, an ROI region filtering sub-module, a ground segmentation sub-module and a recognition sub-module; the position judging sub-module receives the position information of the mine card sent by the mine card positioning system, compares the position information with the position information of the marked position, and sends execution information to the ROI area filtering sub-module when the two pieces of information are consistent; The ROI region filtering submodule receives the execution information sent by the position judging module, reads in the point cloud data of the radar, only reserves the point cloud in the ROI region according to the coordinate value method for limiting the point cloud data, and sends the point cloud to the ground segmentation submodule; the ground segmentation submodule receives the point cloud data sent by the ROI region filtering submodule, obtains ground point cloud and obstacle point cloud data by using a RANSAC method, and sends the obstacle point cloud data to the identification submodule; The identification submodule receives the point cloud data sent by the ground segmentation submodule, calculates the number of point clouds corresponding to the obstacle, obtains the sample characteristics of the operation process, and sends the sample characteristics of the operation process to the performance judging module.
6. The mining vehicle-mounted sensing system performance degradation identification method is characterized in that the prior sample is characterized by the number of point clouds of obstacles in the environment, the vehicle-mounted sensing system comprises a laser radar, the laser radar is used for acquiring environment data, and the number of the point clouds of the obstacles is obtained through processing.
7. The mining vehicle-mounted sensing system performance degradation identification method according to claim 5, wherein the operation process sample is characterized by the number of point clouds of obstacles in the environment.
8. The method for identifying the performance degradation of the mining vehicle-mounted sensing system according to claim 5, wherein the difference between the prior sample characteristic and the operation sample characteristic is an absolute value of a difference between the prior sample characteristic and the operation sample characteristic.

Description

Mining vehicle-mounted perception system performance degradation recognition system and method Technical Field The invention relates to a system and a method for identifying performance degradation of a mining vehicle-mounted sensing system, and belongs to the field of laser radar performance detection. Background The mining area has high dust, high vibration and low temperature scenes, and the laser radar used by the unmanned mining card works under the working condition, so that the problem of performance degradation, such as detection distance and emission point cloud reduction, is easy to occur. And the performance degradation of the radar is not easy to find by directly observing the point cloud data through naked eyes. The performance of the laser radar is reduced, so that the detection capability of a sensing system can be influenced, and the driving safety of the unmanned mining card is further influenced. Prior art solutions require the design of specific devices to detect the lidar and the placement of the lidar in the device. For unmanned mining cards in mining areas, a specific device is designed, the radar is detached, and the detection performance is high in cost and low in efficiency. Disclosure of Invention The invention provides a system and a method for identifying performance degradation of a mining vehicle-mounted sensing system, which solve the problem of the performance degradation of a laser radar for identifying unmanned mining cards in mining areas, which is disclosed in the background art. In order to solve the technical problems, the invention adopts the technical scheme that the mining vehicle-mounted perception system performance degradation recognition system comprises an environment sample feature memory storage module, an environment feature acquisition and recognition module, a performance judgment module and an alarm module; the environment sample feature memory storage module is used for carrying out pre-environment feature acquisition and processing on the marked positions in the environment by utilizing the vehicle-mounted sensing system and storing the results as priori sample features; The environment characteristic acquisition and identification module is used for acquiring operation data in the operation process of the vehicle, acquiring vehicle-mounted sensing data when the vehicle runs to the marked position, and converting the vehicle-mounted sensing data into operation process sample characteristics; the performance judging module is used for receiving the prior sample characteristics from the environment sample characteristic memory storage module and the operation process sample characteristics from the environment characteristic acquisition and identification module, comparing the prior sample characteristics with the operation process sample characteristics to obtain differences between the prior sample characteristics and the operation process sample characteristics, judging that the result is abnormal when the differences exceed a threshold value, and judging that the result is normal otherwise; And the alarm module is used for receiving the judging result of the performance judging module and sending out an early warning signal when the judging result is abnormal. The vehicle-mounted sensing system comprises a laser radar, environmental data are collected by the laser radar, and the number of the point clouds of the obstacle is obtained through processing. Further, the run-time sample is characterized by a number of point clouds of obstacles in the environment. Further, the environmental characteristic acquisition and recognition module comprises a position judgment sub-module, an ROI region filtering sub-module, a ground segmentation sub-module and a recognition sub-module; The position judging sub-module is used for receiving the position information of the mine card sent by the mine card positioning system, comparing the position information with the position information of the marked position, and sending execution information to the ROI area filtering sub-module when the two pieces of information are consistent; The ROI region filtering sub-module is used for receiving the execution information sent by the position judging module, reading in the point cloud data of the radar, only preserving the point cloud in the ROI region according to the coordinate value method of limiting the point cloud data, and sending the point cloud to the ground segmentation sub-module; The ground segmentation submodule is used for receiving the point cloud data sent by the ROI region filtering submodule, obtaining ground point cloud and obstacle point cloud data by using a RANSAC method, and sending the obstacle point cloud data to the identification submodule; The identification sub-module is used for receiving the point cloud data sent by the ground segmentation sub-module, calculating the number of the point clouds corresponding to the obstacle, obtaining the sample cha