Research On Motion Path Planning And Control System Of Fruit Harvesting Manipulator With 7-DOF

Fruit harvesting manipulator is a important equipment for realizing agricultural modernization, so research on the key technologies of control about fruit harvesting manipulator can promote the intelligent level of agriculture infrastructure and improve the level of research and development in agriculture intelligent equipment of our country. This paper took a 7-DOF fruit harvesting manipulator as research object, and the control system, trajectory planning, obstacle avoidance path planning and position sliding mode control of joints of the fruit harvesting manipulator were analyzed and researched respectively.According to the motion features of harvesting manipulator, the harvesting manipulator model was established in Robotics Toolbox, and the theory analysis, and simulation on trajectory planning of third degree polynomial, fifth degree polynomial interpolating function and the method of parabola transition in joints space were accomplished.To accomplish obstacle avoidance motion efficiently for harvesting manipulator,the variational obstacle avoidance method was designed by combining the gradient projection obstacle avoidance method based on pseudo distance index. Simulation results show that the new dynamic obstacle avoidance method accomplished obstacle avoiding task effectively, reduced the total excessive joints movement by 31.7% and single joint displacement by 83.3% in the maximum, and improved the efficiency of obstacle avoidance.In order to research the position sliding mode variable structure control based on exponential approach law of manipulator, the model of single manipulator joint was established on MATLAB. Then the influencement to system dynamic performance and chattering of all parameters of exponential approach law and switch function was analyzed by simulation. Simulation results show that the output chattering of sliding mode control can be reduced by appropriately selecting parameters under the premise of ensuring responsiveness for the system.In order to solve the chattering problem of sliding mode control, the method ofapplying genetic algorithm(GA) to dynamic adjusting sliding mode control parameters were proposed. Genetic algorithm takes the minimum of control output and joint position error as evaluation target, and searches out the optimal sliding mode parameters continuously in real time. The simulation results show that,compare to the common sliding mode control method, the proposed method has stronger robustness characteristics when the system is influencing by external interference, and it can effectively inhibits the chattering phenomenon.According to the research on current mainstream control system architectures of manipulator and characteristics of the research object, this paper presented a distributed control system of main-controller+joints-controller for the manipulator.And aiming at the 6-th joint of harvesting manipulator, hardware of joint control system was designed and made based on STM32 microcontroller and AS5045.Meanwhile, the software for joint position feedback program, position sliding mode control program and send and receive procedures of CAN module were designed.Then the experiment platform for 6-th joint motor was constructed, and the test of position sliding mode control and communication for CAN nodes were conducted on this experiment platform. The experiment results show that although the position sliding mode control of application with system simulation have a gap with a 0.5s longer response time, the position response and CAN communication can fulfill the control requirements of harvesting manipulator.