Rearch On Inter-row Navigation Control Method Of Orchard Mobile Robot Under Low Planting Density Conditions

At present,the low degree of automation of agricultural equipment in China makes agricultural production labor demand and labor intensity.Research on agricultural automation equipment that meets the actual needs of China has been an inevitable need to improve the competitiveness of China’s agriculture.In this paper,we analyze the control and inter-row autonomy problems of wheeled differential steering robots for orchard environment.The working conditions of orchard robots are mostly in outdoor environment,and the robots are affected by weather changes and road undulations when they are in outdoor working conditions.How to make the robots operate autonomously under complicated conditions and have good control accuracy is an important problem for the current orchard mobile robots.This paper compares the basic theory of sliding wheel steering robot(hereinafter referred to as robot)and carries out 3D modeling of the robot according to the theoretical guidance.For the development and verification of the robot autonomous operation algorithm,a multi-body dynamics simulation environment is built and the robot dynamics is selected based on the simulation results.In the robot parameter estimation and control part,this paper proposes a real-time estimation method for the vertical load of the tire of a sliding steering wheeled robot on a non-paved road and an algorithm for real-time estimation and optimal distribution of the tire drive force.In this method,firstly,an ideal plane for the static calculation of the sliding steering process and a four-wheel vertical load estimation method based on this plane are proposed;secondly,a small lateral deflection angle lateral force estimation method based on the FIALA tire dynamics model is proposed;secondly,the steady-state dynamics equations of the sliding steering wheeled robot ramp and the real-time tire drive force estimation method are established;finally,a tire utilization rate based Finally,the optimal real-time distribution model of tire driving force is constructed based on the tire utilization rate.The accuracy of vertical load estimation is more than90%,the accuracy of lateral force estimation is more than 92%,and the tire drive force optimization method based on vertical load and lateral force reduces the tire utilization rate from 96.25% to 93.75%.The paper concludes with an analysis of the robot autonomous operation problem and proposes an outdoor autonomous navigation method based on Li DAR sensors.The inter-row point cloud information is processed using a point cloud processing method,which reduces the number of interfering point clouds and the overall number of point clouds.The processed point clouds are fitted by a random sampling consistent method to generate navigation lines.The actual point cloud is verified that the robot can generate good navigation lines at both large and small heading angles.Finally,the paper analyzes the tracking effect of three trajectory trackers on uneven road surfaces,the results show that the fuzzy controller adapts optimally to the uneven road surface.