Research On Automated Vehicle Motion Planning For Freeway Curves

Lane changing and adaptive following of automated vehicles are the two most common conditions at freeway curves.How to realize the automated vehicle motion planning at freeway curves under all-weather and dynamic conditions is one of the difficulties.In order to make the automated vehicles drive safely,smoothly and efficiently in the curved environment,this thesis will investigate the automated vehicle lane change motion planning and adaptive following motion planning.Firstly,a safe distance model for automated vehicles at freeway curves considering radius of curvature and rainfall is proposed.The lane change and adaptive following motion planning models are prepared for the construction of the model,so that the constructed model can plan a safe,reasonable and stable path and speed profile.Secondly,when the car in front of the same lane on the freeway drives at low speed for a long time in the curve and the road section allows lane change,the vehicle ahead brakes urgently or there is a stationary obstacle in front,the main car adopts the strategy of lane change.The lane change path planning and speed planning are decoupled to study the lane change motion planning model of automated vehicles at freeway curves.A cluster of lane change path curves is generated based on cubic B-sample curves,and then the optimal lane change path curve is filtered based on the multi-objective cost function.Moreover,based on the optimal lane change path,speed planning is carried out by first mapping all surrounding vehicles into the S-T diagram,searching the S-T curve cluster based on the improved dynamic planning algorithm under the speed and acceleration constraints,and then obtaining the optimal S-T curve based on the cost function and quadratic planning optimization,and then deriving the optimal speed curve.Last,the optimal lane change path curve and the optimal velocity curve are combined to obtain the optimal lane change motion planning for automated vehicles at freeway curves.Furthermore,the main vehicle adopts an adaptive following strategy when following a vehicle at a freeway curve.A motion planning model with hierarchical adaptive following path planning and speed planning at freeway curves is proposed,and the adaptive following path is planned based on the improved artificial potential field method to make the main vehicle drive on the centerline of the lane as much as possible.Then the speed planner is designed based on this adaptive following path,and finally the adaptive following path is combined with the speed curve,which is the adaptive following motion planning of the automated vehicle at freeway curves.At last,the adaptive following path is combined with the speed curve,which is the adaptive following motion planning of the automated vehicle in the freeway curve.Finally,a simulation platform is built to verify the feasibility of the motion planning model proposed in this thesis.A vehicle control model based on model predictive control is established,and a joint simulation platform is built to conduct simulation experiments on lane changing and adaptive following of automated vehicles at freeway curves.The simulation results show that the minimum safe distance model can improve the safety of the main vehicle motion planning;the lane changing and adaptive following motion planning model can ensure the safety,stability and efficiency of the driving process;the motion tracking control model has good tracking effect and small tracking error.