Research Of Self-propelled And Intelligent Platform In Orchard

Orchard industry progresses day by day,however,corresponding weeding systems are either semi-manual style,or intelligent systems with high cost,high difficult design and operation still on the stage of theoretical research.Aiming at the above phenomenon,in this paper,an orchard intelligent system which was simple and easy to popularize was designed.This system can help the orchard farmers to enhance work efficiency,improve annual output,reduce manual labor and avoid environmental pollution,and also can promote the popularization of intelligent orchard.The intellectualization of this system mainly lies in:Self-propelled navigation,obstacle avoidance between the lines,inline obstacle avoidance and independent turn.(1)The design of the mechanical body of the mobile platform was completed,the design targets and requirements was formulated.The wheel configuration scheme,the type and the driving mode of the wheel were determined,the steering mechanism and the anti-collision beam mechanism were also designed.(2)In order to make machine comprehensively intelligent,the autonomous navigation system based on GPS was installed.In order to improve the positioning accuracy,in addition to using the differential positioning system,the kalman filtering and the PD control system for steering were used.The experimental results showed that when the mobile platform runs straight line navigation,the maximum lateral deviation was 0.141m,with a mean of 0.070m;when the mobile platform run turn navigation,the maximum lateral deviation was 0.227m,with a mean of 0.092m.The navigation system of the mobile platform can basically realize real-time traveling according to the default routes,meeting the demands of this topic.(3)A kinematical analysis was carried out on the wheel-type intelligent mobile platform,and the final motion model of wheeled mobile platform in the world coordinate system was deduced,and also the navigation path of the platform when working in the orchard was planned.(4)According to the actual situation of weeding platform,the most suitable motor type and motor model were selected,and MATLAB software was used to make simulation analysis of the transient process of the motor.The simulation results showed that under the condition of high speed and low torque,the control mode of MTPA and Id=0 both could make the motor reach steady state with the smaller torque ripple.When the weeding mobile platform under special working conditions,MTPA control was better than Id=0 control and the motor output torque response was fast,and under the same output torque,electric current was smaller,with less power consumption.(5)Inter-row navigation obstacle avoidance radar was for other obstacles,and compared feedback signals of three symmetric symmetrical ultrasonic sensors at front end to plan route for obstacle avoidance.Best measured in experiment and row distance of 2m of obstacle avoidance,obstacle avoidance rate of 90%;the average time of obstacle avoidance for 10.6s;the biggest obstacle avoidance during lateral deviation is 1.132m;the average stability after lateral deviation error is 0.078m.(6)Software design of the navigation system was completed,and the human-computer interaction interface was created.Through the receiving and analysis of GPS signal data we completed GPS positioning information extraction and coordinate transformation.According to the straight and steering navigation control methods we completed the compiling of the whole navigation system program,realizing intelligent walk straight and steering function of the robot.