Greenhouse Autonomous Operation Platform Based On Laser Slam Design And Test Of Navigation System

It has changed the traditional mode of agricultural production and the development direction of modern agriculture.The greenhouse operation environment is narrow,the conventional field operation equipment is difficult to apply,the labor intensity of greenhouse production operation is high,and the key operation links such as manual spraying of chemical pesticides will cause harm to people.It is very necessary to study the intelligent autonomous navigation operation platform suitable for the greenhouse environment.Environment awareness is the basis of autonomous navigation application of operation platform and the prerequisite of realizing autonomous navigation.The greenhouse environment is closed,the satellite signal is seriously blocked,and the satellite navigation signal is unreliable.Exploring the positioning mode that does not depend on satellite navigation is an important way of greenhouse environment navigation technology.Simultaneous localization and mapping(SLAM)based on laser has the characteristics of strong robustness and wide application range,and can realize the map construction of closed scenes.However,there are phenomena such as lack of seedlings and sparsity in greenhouse crop planting,which makes the map boundary unable to be closed and difficult to navigate directly.By exploring the impact of unstructured greenhouse environment on slam and autonomous navigation,this paper studies the navigation method of sparse crop rows in Greenhouse Based on laser slam;Integrating the information of lidar,inertial measurement unit(IMU)and encoder,using A* algorithm and dynamic window approach(DWA)algorithm,a slam navigation system of greenhouse autonomous operation platform based on multi-layer cost map is designed to realize the mapping and autonomous navigation of sparse crop rows in warm room environment.The main research contents of this paper are as follows:(1)Research and design the greenhouse autonomous operation platform and autonomous navigation software and hardware system.According to the greenhouse planting mode and the distribution characteristics of obstacles,the greenhouse autonomous operation platform and navigation software and hardware system are designed by using the method of single line lidar,encoder and IMU multi-sensor fusion,so as to realize the greenhouse environment slam and autonomous navigation operation.(2)Differential kinematics model,track deduction model and navigation control system of autonomous operation platform are studied and designed.The laser slam and global or local path planning algorithms are compared and analyzed to determine the best application scheme suitable for greenhouse working environment;A slam navigation optimization method for greenhouse crop canopy based on multi-layer cost map is proposed.The cost map is divided into static layer,obstacle layer,expansion layer and virtual crop layer,which can effectively avoid the problem of inaccurate navigation path planning and even planning failure caused by the degradation of slam map in unstructured environment and sparse crop rows.(3)The slam navigation test results show that the navigation system of the autonomous operation platform can meet the needs of slam and navigation in a variety of environments.When the autonomous operation platform operates at the speed of 0.2,0.5 and 0.8m/s respectively,the average deviation distance between the actual navigation position and attitude and the navigation target point in the laboratory environment is less than 6cm,and the heading deviation is less than 5°;In the actual greenhouse environment,the average deviation distance between the actual navigation point and the navigation target point is less than 15 cm,and the heading deviation is less than 18°,meeting the accuracy requirements of the actual navigation operation in the greenhouse environment.This research can provide technical support for the research and development of greenhouse robot navigation system.