Study On Path Planning Of Orchard Mower Robot Based On Lidar

The fruit tree planting area and fruit production in China have been increasing year by year,which has led to an increase in the workload of orchard management.For some orchards,Such as litchi and longan orchards,in the work of weeding,it mainly depends on whether the orchard managers use the hanging mower or hand-plow mower on the back.In order to improve the level of automation in orchard mowing operations,save labor and promote the application of science and technology to agricultural machinery.This paper designs and researches an orchard mowing robot based on lidar,designs the driving control system and path planning method on the existing mower platform,and realizes the autonomous walking under simulated orchard environment.The main research contents of this article are as follows:1)Design and build a driving control system for mowing robots.Analyze the characteristics of the orchard environment to obtain the functional requirements of the mower,design the framework of the control system,and select the type of hardware and corresponding control program for the sensor.Finally set up a lidar mower walking control system based on lidar.2)Design a road roughness detection mechanism.Because the orchard pavement is bumpy and prevents equipment from accidents such as rollover,a set of road roughness testing mechanism was designed based on the mechanical structural parameters of the mower body and road conditions of the orchard.And for these pavement design corresponding walking path control strategy.3)Determine the path planning plan of the mowing robot in this study.By analyzing the path planning methods of mobile robots and the characteristics of the orchard environment,the designed solutions are combined with cell decomposition method and traversal method based on sensor local path planning.The cell decomposition method is mainly used to break down the orchard area into some non-coincident units when there are relatively large obstacles or pits and bumps in the orchard.Sensor-based local path planningis mainly based on the perception of the environment by sensors such as laser radar to realize the traversal of mowing areas.4)Using the mowing robot platform to test.The experimental results show that Lidar can identify and extract the location information of the orchard boundary and end point.In the electronic compass-based walking test,the average deviation of the straight walking was 1.76°,and the average deviations in the correction tests of the heading angles of 10°,20°,and 30°were 1.56°,1.8°,and 1.84°,respectively.In the row-turning test with row spacing of 90 cm,the average deviation of the angle is 1.98°,the average deviation of the line-lapping distance is 5.4cm,and the maximum deviation is 9cm.The average deviation of the angle of the electronic compass basically meets the measurement error range of 2°,but in a single experiment,individual test results exceed this measurement error,and the maximum deviation is 3.8°.In general,actions such as straight-line walking and turning around can meet the requirements.In the scheme of path planning,it is possible to implement traversal mowing of grass cutting areas.