| With the continuous expansion of orchard planting area and the rapid development of autonomous and intelligent orchard agricultural equipment,orchard robots have been widely used in many fields such as inspection,spraying,weeding,picking and handling.The orchard autonomous mobile robot can carry out autonomous motion planning according to the known environmental map,realize autonomous navigation and intelligent operation in combination with operation equipment with different functions,reduce labor costs and improve operation efficiency.In order to improve the autonomy and operation efficiency of mobile robots in orchards,this paper studies the motion planning methods for autonomous obstacle avoidance of mobile robots in orchards.planning method.The main research contents are as follows:(1)By analyzing the functional requirements of the orchard mobile robot autonomous obstacle avoidance motion planning system,the overall design of the mobile robot autonomous obstacle avoidance motion planning system from both software and hardware aspects has been completed,and the principle analysis and selection of key hardware modules have been completed.And design and build a mobile robot test platform,based on the ROS platform to design the software algorithm framework.(2)According to the characteristics of the orchard environment and the needs of motion planning,the airborne two-dimensional lidar is used to collect the position information of fruit tree trunks,and combined with the real-time pose information of the mobile robot fed back by the global positioning module,the position information of the fruit trees is converted into the global coordinate system,and The fruit trees and obstacles are inflated,the starting and ending points are determined,the boundaries of the orchard are established,and the global geometric feature map of the orchard environment is constructed as the prior information of the autonomous obstacle avoidance motion planning system.(3)Aiming at the problems of the traditional RRT* path planning algorithm in the continuous corridor environment,such as low search efficiency,low sampling point utilization,and large multi-turning angles of the generated path polylines,an improved bidirectional RRT* mobile robot motion planning algorithm is proposed.In the global geometric feature map of the orchard,the improved bidirectional RRT* algorithm is used to search for paths,combined with dynamic terminal node guidance and potential field guidance for bias sampling,and the initially generated paths are subjected to path point de-redundancy and adjacent polyline segment corner constraints.;Finally,a third-order quasi-uniform B-spline curve is used to optimize the trajectory of the processed path points.The experimental results show that compared with the traditional bidirectional RRT* algorithm,the improved algorithm in this paper reduces the planning time by57.5% on average,the sampling point utilization rate is increased by 30.05% on average,and the final planned path is shortened by 7.14% on average.(4)A local real-time trajectory re-planning method based on global guide lines is proposed for the situation of orchard environment changes and dynamic obstacles.Based on the Frenet coordinate system,this method plans the horizontal and vertical trajectories separately for different scenarios that may appear in the orchard,merges the generated horizontal and vertical trajectories,and designs a quality evaluation function for the synthetic trajectory to filter out the trajectories that do not meet the kinematic constraints.Then,the trajectory with the smallest cost function value is selected as the optimal trajectory.Finally,an orchard is simulated to build a simulation environment,which verifies the applicability and safety of the method in the orchard environment. |
PDF Full Text Request