Design And Application Research Of 3D Environmental Scanning System For Electric Shovel

Electric shovel is the main mining equipment for large open pit mining.At present,electric shovel is mainly operated by human,which has the problems of low full bucket rate,low loading efficiency and high energy consumption.It is an effective way to solve the above problems to realize the automatic excavation of shovel by self-planning excavation trajectory,automatic excavation and automatic loading technology.In order to realize the automatic excavation of the electric shovel,it is necessary to obtain the three-dimensional topographical information of the excavation site online.Therefore,it is of great significance to study the threedimensional environmental scanning system applicable to the electric shovel.In view of the requirements of the three-dimensional environmental scanning system of electric shovel,this paper proposes a design scheme of three-dimensional environmental scanning system of shovel with low cost,wide measuring range and strong adaptability to working conditions and corresponding three-dimensional point cloud data processing method.According to the design scheme,the hardware construction and software algorithm development of the system are completed on the experimental platform of Taizhong WK-55 shovel prototype.According to the actual needs,the data acquisition scheme is designed,including the scanning scheme of 2D lidar,the positioning and attitude detection scheme of dual GNSS antenna based on RTK,the data filtering and attitude resolution scheme of inertial navigation sensor.According to the output data characteristics of the data acquisition system,the embedded system scheme for data acquisition and fusion is designed.The hardware part completes the communication scheme design,power supply system design,data processing system circuit design,anti-interference and protection processing of the hardware system.The program is designed to receive,parse and fuse the data of lidar,RTK and INS/satellite mobile station.The real-time and high-precision position and attitude information of shovel and the lidar data with GNSS timestamp can be obtained by the program processing.According to the data sent by the embedded system to the host computer,the host computer program development is first completed to realize the receiving,storing and visualizing of the data.A minimum time difference matching algorithm is proposed to achieve the best matching of the 2D lidar data and the horizontal shovel angle of the shovel in time to fuse the 3D point cloud data into the environment.The accuracy of the acquired 3D point cloud number is analyzed and verified by experiments.Converting the acquired 3D point cloud data into the coordinate system of the shovel experimental platform by the rotation matrix method.The Kd-tree data structure is established for the point cloud data in the experimental platform coordinate system,and the point cloud data is filtered based on this.The maximum connected region algorithm and the Gaussian filtering algorithm are used to remove outliers in the point cloud data.Finally,based on the edge feature of the point cloud data,the bilateral filtering algorithm is used to correct the edge points.This paper completes the design and construction of the three-dimensional environmental scanning system of electric shovel,and carries out experimental verification,and carries out effective post-processing of the acquired data.Three-dimensional environmental scanning system can accurately measure the attitude of shovel equipment through two complementary attitude measurement systems and filtering algorithm.The yaw angle accuracy can reach 0.1 degree.The position accuracy of shovel can reach 2 cm through RTK-GNSS measurement scheme.The minimum time difference matching algorithm can match the point cloud data and yaw angle of 2D lidar in time to obtain the three-dimensional point transport data of the environment.The time matching error is less than 8 ms.The accuracy of point cloud data measured in the range of 10 m is less than 8cm,which meets the requirement of Engineering application.The quality of point cloud data can be effectively improved by indexing and filtering the point cloud data.