Research On General Mobile Operating Platform In Greenhouse

How to realize the independent walking of the operating platform is the key technical problem in the design of the universal mobile platform in the greenhouse.The thesis aims at realizing the independent movement of the operating platform in the greenhouse environment.It mainly carries out the construction of the platform of the mobile job test,the construction of the work environment map,the theory research,the simulation analysis and the laboratory test of the robot location and the job path planning.The working environment of the mobile platform--the actual situation of the greenhouse is analyzed,and the general mobile operation testing platform in the greenhouse is built.Considering the accuracy of the turning angle and the load bearing capacity,the mobile platform uses three wheel structures,the front two wheel differential speed control turns,the rear wheel is used as the moving wheel,and the laser radar and ultrasonic sensors are used to obtain the environmental information.Considering the expansion of the system,the software development environment of the mobile platform chooses the ROS operating system as the upper computer software platform to realize map creation and path planning.The lower computer uses 32 bit STM chip processor to realize the motion control of the platform.In the construction of mobile platform map,the laser radar sensor is used to obtain the environmental information.A method of building a grid environment map based on Rao-Blackwallised particle filter is proposed.The environment map is built in the actual environment,and the effectiveness of the algorithm is verified.Experiments show that map building algorithm based on Rao-Blackwellized particle filter can get a map with high accuracy and consistency.The basic principle of the global path planning algorithm A* algorithm and the local path planning algorithm DWA algorithm is analyzed and studied on the navigation and positioning of the mobile platform.On the basis of the research of the A* and DWA algorithms,the simulation experiments of the single target point planning,the square path moving and the independent obstacle avoidance are carried out on the basis of the research of the A* and DWA algorithms,and the validity of the algorithm is verified.In view of the tracking and positioning problem,the AMCL algorithm theory is studied.The simulation experiment results show that the number of particles can be dynamically adjusted by the KLD sampling method,and a relatively better positioning effect is obtained with fewer particles.Finally,in the actual environment,combined with the current environmental information obtained by lidar sensors,we conduct straight path navigation,bend navigation and obstacle avoidance navigation experiments.The experimental results show that the mobile platform can walk independently through the path planning algorithm,and can analyze the obstacle information in the environment and plan the real time obstacle avoidance path.