CN-122017871-A - Unmanned mine car loading and parking space positioning method based on laser radar point cloud

CN122017871ACN 122017871 ACN122017871 ACN 122017871ACN-122017871-A

Abstract

The invention discloses a laser radar point cloud-based unmanned mine car loading parking space positioning method, which comprises the steps of obtaining laser radar point cloud data comprising an excavator and mine car environments, preprocessing the laser radar point cloud data, establishing a target pose model of a loading parking space based on safety standards, carrying out point cloud registration on a current point cloud corresponding to the current position of a mine car and the target pose model, calculating relative pose deviation between the current pose and the target pose of the mine car, calculating pose adjustment quantity required by the unmanned mine car to drive into the loading parking space based on the relative pose deviation, and controlling the unmanned mine car to move to the loading parking space according to the pose adjustment quantity. According to the method, the laser radar point cloud is processed, the target berth model is established, the target berth model is accurately registered with the current point cloud of the mine car, and the mine car is controlled to automatically stop into a loading berth after pose deviation is calculated. Realize unmanned mine car loading berth's full-automatic, high accuracy location, promoted operation security, continuity and overall efficiency.

Inventors

DU YANCHUN
CHEN CHENG
XU WENTAO
WANG GUIDONG

Assignees

安徽海博智能科技有限责任公司
安徽海螺集团有限责任公司
安徽海螺水泥股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260227

Claims (10)

1. The unmanned mine car loading parking space positioning method based on laser radar point cloud is characterized by comprising the following steps of: S1, acquiring laser radar point cloud data comprising environments of an excavator and a mine car; S2, preprocessing the laser radar point cloud data, and establishing a target pose model of the loaded parking space based on safety specifications; s3, carrying out point cloud registration on the current point cloud corresponding to the current position of the mine car and the target pose model, and calculating the relative pose deviation between the current pose of the mine car and the target pose; S4, calculating the posture adjustment quantity required by the unmanned mine car to drive into a loading parking space based on the relative posture deviation; And S5, controlling the unmanned mine car to move to the loading parking space according to the posture adjustment quantity.
2. The method for locating the loading parking space of the unmanned mine car based on the laser radar point cloud according to claim 1, wherein in the step S2, the establishing the target pose model specifically comprises: and establishing a global parking coordinate system by taking the center of the excavator bucket as an origin, extracting a geometric outline of a berth boundary through clustering segmentation based on the processed point cloud data, and further generating the target pose model.
3. The method for locating the loading parking space of the unmanned mine car based on the laser radar point cloud according to claim 2, wherein in the step S2, the laser radar point cloud data is preprocessed, and the method specifically comprises the following steps: and splicing the point clouds acquired by the multiple sensors or the multiple frames to be unified to the same coordinate system, and denoising and smoothing the point clouds.
4. The method for locating the loading parking space of the unmanned mine car based on the laser radar point cloud as set forth in claim 1, wherein the step S3 specifically includes: s31, dividing the current point cloud into a mine car body point cloud and an excavator point cloud; step S32, registering based on the excavator point cloud and the corresponding excavator reference point cloud in the target pose model to obtain initial pose transformation parameters; And step 33, transforming the mine car body point cloud by using the initial pose transformation parameters, and performing rough registration and fine registration on the transformed mine car body point cloud and the target pose model to calculate the relative pose deviation.
5. The method for locating the loading parking space of the unmanned mine car based on the laser radar point cloud as recited in claim 4, wherein in the step S31, the point cloud is divided into the mine car body point cloud and the excavator point cloud through a preset division plane equation.
6. The method for locating the loading parking space of the unmanned mine car based on the laser radar point cloud according to claim 4, wherein in the step S32, an iterative nearest point algorithm is adopted to register the excavator point cloud with the excavator reference point cloud, and a rotation matrix and a translation vector are obtained as the initial pose transformation parameters.
7. The method for locating a loading parking space of an unmanned mine car based on a laser radar point cloud as recited in claim 4, wherein in the step S33, the coarse registration is performed by using a sampling consistency initial registration algorithm, and the fine registration is performed by using an iterative closest point algorithm.
8. The unmanned mine car loading parking space positioning method based on laser radar point cloud as claimed in claim 7, wherein in the fine registration process, point cloud data of an excavator cab are extracted from the target pose model to serve as registered target point clouds.
9. The method for positioning the loading parking space of the unmanned mine car based on the laser radar point cloud as recited in claim 1, wherein in the step S4, the attitude adjustment amount includes a deviation of distances between the center of the mine car and the center of the parking space in the transverse direction and the longitudinal direction, and an included angle deviation of a fitting straight line between a heading angle of the mine car and a boundary of the parking space.
10. The method for positioning the loading and parking space of the unmanned mine car based on the laser radar point cloud as claimed in claim 1, wherein in the step S5, a control command is generated to dynamically control the mine car to travel to the loading and parking space according to the attitude adjustment amount and in combination with the mine car kinematic model and the steering constraint.

Description

Unmanned mine car loading and parking space positioning method based on laser radar point cloud Technical Field The invention relates to the technical field of supervision information fusion of surface mines, in particular to a laser radar point cloud-based unmanned mine car loading parking space positioning method. Background The loading parking mainly refers to that in a mine running area, an unmanned mine car transmits information such as a fixed shovel position, an orientation angle and the like according to an excavator, the information such as the type of the excavator and the shovel position which need to be adapted is judged, and a shovel area path is planned. The device can be effectively, stably and orderly parked to the corresponding shovel position from the tracking end point. After the loading is completed, the loading can be stably parked and reach a parking end point appointed by a main road, and the handover can be smoothly completed for the whole loading parking Comprehensive analysis of the optimal parking point locating method of the unmanned mine car is combined with mining area operation characteristics and technology to realize paths, and the method is divided into three main stream methods: (1) Collaborative calculation method based on excavator positioning And calculating the parking point of the mine car through the real-time position of the excavator. The method has the advantages of dependence on the existing equipment and strong real-time performance. The limitation is that high-precision GPS support is needed, and complex terrains are easy to be interfered by signals. (2) Grid map and track optimization method Generating candidate parking spaces based on environmental modeling, and screening an optimal path: The method has the advantages of being suitable for complex obstacle environments and high in path planning success rate. The method has the limitation that the implementation complexity is high, and the grid map is not suitable for a large-scale environment, a dynamic scene and a resource-limited system. (3) Key control point guidance method Parking accuracy is controlled through the path key nodes: The method has the advantage of solving the problem of long-distance reversing deviation. The method is particularly suitable for scenes in which the mining point is far away from the main road operation area, and other scenes have unobvious advantages. In order to solve the problem of low locating efficiency of the optimal charging and parking point of the unmanned mine car, a laser radar-based locating method of the optimal charging and parking point of the unmanned mine car is needed. Disclosure of Invention The invention aims to overcome the defects in the prior art, and aims to solve the problems in the prior art by adopting a laser radar point cloud-based unmanned mine car loading parking space positioning method. A laser radar point cloud-based unmanned mine car loading parking space positioning method comprises the following steps: S1, acquiring laser radar point cloud data comprising environments of an excavator and a mine car; S2, preprocessing the laser radar point cloud data, and establishing a target pose model of the loaded parking space based on safety specifications; s3, carrying out point cloud registration on the current point cloud corresponding to the current position of the mine car and the target pose model, and calculating the relative pose deviation between the current pose of the mine car and the target pose; S4, calculating the posture adjustment quantity required by the unmanned mine car to drive into a loading parking space based on the relative posture deviation; And S5, controlling the unmanned mine car to move to the loading parking space according to the posture adjustment quantity. In the step S2, the establishing the target pose model specifically includes: and establishing a global parking coordinate system by taking the center of the excavator bucket as an origin, extracting a geometric outline of a berth boundary through clustering segmentation based on the processed point cloud data, and further generating the target pose model. In the step S2, preprocessing the laser radar point cloud data specifically includes: and splicing the point clouds acquired by the multiple sensors or the multiple frames to be unified to the same coordinate system, and denoising and smoothing the point clouds. As a further scheme of the present invention, the step S3 specifically includes: s31, dividing the current point cloud into a mine car body point cloud and an excavator point cloud; step S32, registering based on the excavator point cloud and the corresponding excavator reference point cloud in the target pose model to obtain initial pose transformation parameters; And step 33, transforming the mine car body point cloud by using the initial pose transformation parameters, and performing rough registration and fine registration on the transformed mine car body