Research On Path Planning And Motion Control Of Unmanned Ground Vehicle On Unstructured Road

In recent years,unmanned ground platforms have been widely used in material transportation,terrain survey,outdoor rescue and other tasks by virtue of their advantages of good power and mobility in complex structural environments.In order to further improve the efficiency of the unmanned platform and reduce the input of labor cost,higher requirements are put forward for the unmanned platform to achieve complete autonomy on the unstructured road.However,different from structured scenes such as urban flat road,unstructured environment has uneven ground,complicated terrain,and no clear road boundary,which brings great challenges to autonomous driving of ground unmanned platform,especially in path planning and motion control.However,the current unmanned driving technology is mainly oriented to the driving environment of paved road,which is difficult to directly transplant to the unstructured road.Therefore,combined with a research project,this paper takes the sixwheeled unmanned platform as the research object to study the path planning method and motion control method under the unstructured road.The main research contents include the following aspects:Firstly,the capacity of unstructured road is quantified.Firstly,the key indexes to evaluate the capacity of unmanned platform on unstructured roads are studied from the perspective of terrain geometric features.Then,the obstacle crossing ability of the unmanned platform in typical unstructured terrain is studied and the ultimate obstacle crossing performance parameters of the unmanned platform are calculated.Based on the traffic capacity evaluation Index and limiting obstacle crossing performance parameters,the IOP(Index of Passibility)traffic capacity model was proposed to comprehensively quantify the traffic capacity of the unmanned platform in the unstructured environment.Finally,the IOP raster map is built based on the IOP capacity model and elevation model information.Secondly,the path planning problem of unmanned platform under unstructured road is studied.An adaptive heuristic function based on IOP traffic capacity model is designed,and an improved A* path planning algorithm is proposed for global path planning.Combining dynamics model constraints and grid boundary constraints,local trajectory planning was carried out by solving optimization problems.Finally,the global path planning simulation and local trajectory planning simulation are carried out on the IOP raster map constructed in the experiment.Thirdly,the motion control problem of unmanned platform with complex structure is studied.The six-wheel unmanned platform tire model,wheel dynamics model and vehicle body dynamics model under compound ramp were established.The extended Kalman filter was used to estimate the transverse and longitudinal slope angles of the unmanned platform.Then,the error tracking model of the unmanned platform was established.The lateral error,course Angle error and their change rate were taken as the state variables,and the expected yaw of the vehicle body was taken as the control variables.The model predictive control lateral controller was designed.The control rate was designed based on the expected total driving force and the longitudinal velocity error.The sliding mode control method was used for the path tracking longitudinal control,and the stability of the sliding mode controller was analyzed.Fourthly,the path planning and motion control algorithm of the unmanned platform are simulated and verified.Firstly,the torque distribution control strategy is designed based on the principle of minimum sum of tire load rate and considering the non-skid control of unmanned platform drive.A comprehensive simulation platform for motion control of the six-wheeled unmanned platform was built by Simulink,including environment map module,path planning module,path tracking control module and vehicle dynamics model module.At the same time,the approximate double shift condition and approximate straight line were selected to simulate the path tracking control.The rationality and effectiveness of the motion control algorithm and torque distribution control strategy were verified by quantitative comparative analysis of the error indexes in the simulation results.The key technologies of path planning and motion control of unmanned platform on unstructured road are explored through the above-mentioned researches on the quantification of traffic capacity of unmanned platform,path planning of unmanned platform on unstructured road,motion control of unmanned platform on complex structure,and path planning and motion control algorithm simulation verification,which lays a theoretical foundation for autonomous research of unmanned platform.