Research On Motion Planning Of Driverless Articulated Sweeper

Driverless sweeper can be used in parks,communities,factories and other scenes,so the driverless sweeper has a broad prospect in engineering applications.Based on the real vehicle platform of articulated driverless sweeper provided by an enterprise,designed a driverless scheme for the functions of articulated driverless sweeper,such as path planning,following,constant speed cruising,following,lane changing,autonomous overtaking,intersection traffic,active obstacle avoidance,red and green light start and stop,automatic start and stop,braking and fixed-point parking.The software and hardware platform architecture is built,and the trajectory planning and speed planning are studied and designed for the key technology of the unmanned sweeper.The research work of this paper is as follows.(1)Trajectory planning research,in view of the unmanned articulated sweeper kinematics equation is more complex,more constraints,this paper combined with the articulated sweeper kinematics model,through the curvature cost,hinge angle change rate cost combined with the fifth order Bessel curve to generate a large number of curvature continuous and meet the hinge angle constraint candidate trajectory.At the same time,considering the working conditions of the sweeper,the anti-collision constraints of the sweeper are established,the front vehicle trajectory is substituted into the cost function of the sweeper as a constraint to cope with the dynamic changes of the environment.The minimum cost trajectory is selected by cost function,and the optimal trajectory is obtained as the output of motion planning.(2)Research on speed planning,aiming at the problem of "overshoot" of driverless articulated sweeper,this paper analyzes the speed constraints of driverless articulated sweeper,finds the relationship between speed and articulated angular speed,and ensures the stable steering of the sweeper in the normal working process.Then,the velocity planning based on the target distance is designed to ensure the accurate behavior of the sweeper on the whole reference trajectory.At the same time,aiming at the dynamic obstacles,this paper analyzes the working conditions of the sweeper without speed planning,so as to ensure that the sweeper can calculate the acceleration in real time according to the distance of the obstacles and the current speed,and complete the acceleration and deceleration process safely and smoothly.(3)Finally,based on the platform of a sweeper,the motion planning method of articulated driverless sweeper under various working conditions is verified.The experimental results show that the trajectory generated by the algorithm is reliable and real-time,which can ensure the fast obstacle avoidance and stable acceleration and deceleration of the sweeper,and improve the efficiency of the sweeper.