Research On Global Path Planning Of Autonomous Mobile Robot In Strawberry Greenhouse Based On ROS

As a kind of high-value economic crop,strawberry is currently planted in large areas all over the country.The environmental climate in different regions is inconsistent.Therefore,most places have constructed various strawberry greenhouses for planting according to local conditions.The robot can undertake inspection,transportation and picking in the strawberry greenhouse,which can reduce labor costs and improve production efficiency.However,according to the task requirements,it is difficult for the robot to adapt to multiple environments.Autonomously planning the optimal path and then autonomously navigating to the designated position of the task is also a key problem that the robot needs to solve in the application of the strawberry greenhouse greenhouse environment.In view of the above problems,the work done in this thesis is as follows: Selecting autonomous mobile robot as the research platform;According to the precision required by the application,the sensor that obtains the environmental depth information is modified.According to the complex strawberry greenhouse with high degree of freedom of moving road and large area,the global path planning algorithm in the robot navigation function package is mainly studied in.Hardware and hardware modification: This thesis uses the XQ-mini Xiaoqiang robot of Blue Whale Intelligent Robot Company as the research platform.Under consideration of reducing the cost of robot equipment as much as possible,Kinect,whose price is lower than that of laser radar but whose precision meets the requirement of the task,is selected as the sensor for acquiring the environmental depth information.At the same time,in order to enable the obtained information to be used in the gmapping function package,the information is converted;By establishing the coordinate relationship between the robot and Kinect,the strawberry greenhouse environment and the robot position relationship are constructed.Algorithm improvement: Firstly,the reason of Dijkstra algorithm’s low efficiency in path search in strawberry greenhouse with high degree of freedom of moving road and large area is analyzed.Next,the scheme of using the best first search algorithm to improve the efficiency of path search is simulated.Analysis shows that although the algorithm can improve the efficiency of path search,it may not be able to search the shortest path.Then,in order to improve the efficiency of path search on the premise of finding the shortest path,the A~* algorithm is chosen to search for the path.The problems of "non-optimal path"," path undersmooth " and " path search inefficiency " in the use of A~* algorithm are analyzed and improved respectively,and the final improved algorithm is simulated by MATLAB.The results show that compared with A~* algorithm,the improved algorithm is more efficient in searching path,and the path is shorter and smoother.Finally,the improved algorithm is used to replace the global path planning algorithm in the robot’s original navigation function package,and the experimental verification proves that the improved path planning algorithm implemented by the modified Xiaoqiang robot in this thesis can optimize path planning in different strawberry greenhouse environments and autonomously navigate to the designated position of the task.Improving the path search efficiency of the robot in strawberry greenhouse environment can improve the work efficiency of strawberry greenhouse planting,reduce labor costs and increase productivity.At the same time,the research results of this work can not only provide some ideas and methods for the robot application in other similar agricultural production scenes,but also provide ideas for the robot path optimization and autonomous navigation in complex environment.