Research On Autonomous Path Planning Algorithm Of Lawn Trimming Robot Based On Machine Vision

With the development of the theories on the intelligent control,artificial intelligence and so on,the research and application of mobile robot technology have attracted the attention of scientific research enthusiasts all over the world.The wide application of lawn greening machinery facilitates people's life,but automatic and intelligent greening equipment started late in our country.In view of this situation,we have studied the lawn mowing robot.We proposed to build a lawn map through the robot's automatic edge finding,and then pre-plan the lawn mowing path based on the autonomous operation of the equipment.In the process of pruning according to the pre-planned path,we have explored the relevant key technologies by detecting obstacles and avoiding obstacles in real time.The main contents are as follows:Firstly,the hardware design of lawn mowing robot control system is completed,including navigation and positioning device,external environment detection device,machine vision device,etc.Navigation and positioning used the mode of combining GPS and gyroscope to obtain the position information of the robot,and external environment detection adopted the mode of combining lidar sensor and ultrasonic to collect the distance of lawn obstacles.A monocular vision camera is used to collect lawn images,then we determined STM32ZF407 to control the action of the robot as the master controller.Secondly,the identification algorithm of the lawn contour is studied.Based on the images obtained by the visual camera,the lawn boundary is identified,the robot's automatic edge finding is completed.The lawn boundary path is linearly fitted to complete the extraction of the lawn boundary.And the lawn contour map is constructed by combining the positions obtained by GPS and gyroscope.Thirdly,the autonomous path planning algorithm of robot is studied.Based on the operation path of lawn outline planning map robot,we improved the research of the A*algorithm.The improved algorithm works more efficiently in special environment maps.The hybrid planning method of global path planning and local path planning of artificial potential field algorithm is used to realize the autonomous path planning of the lawn mowing robot.Finally,the obstacle avoidance algorithm in the operation process of the robot is studied,and we established the robot motion model firstly,the robot self-positioning is realized through coordinate transformation.In order to using sensors to obtain obstacle contour information aiming at the obstacles that the lawn mowing robot may encounter in the working environment,A real-time obstacle avoidance algorithm is designed.