Methods For Trajectory Planning And Tracking Control Of Buses Autonomously Moving In And Out Of Harbor-Shaped Stops

The non-regulated stopping and free stopping of buses occurred in many cities,which lead to increased traffic congestion and frequent traffic accidents of bus stops.The normal operation of bus routes and the safety of passengers travelling are seriously affected.It is a major challenge for conventional public transport system to ensure that buses can complete the stopping process safely,efficiently and regularly.With the rise of intelligent vehicles,conventional public transport is also gradually moving towards intelligence.As the two key technologies,trajectory planning and tracking control can effectively improve the safety and efficiency of buses.It becomes the important issues in the field of intelligent transportation.Thus,in order to promote the intelligent development of conventional public transport,reduce the incidence of traffic accidents in the area of bus stops and ensure the safety,and economy of urban residential travel,the paper takes the harbor-shaped stop with complex environment as the subject and provides an in-depth study of the methods for trajectory planning and tracking control of buses autonomously moving in and out of stops.Firstly,the basic functions,setting methods,setting locations and design parameters of harbor-shaped stops are sorted out.On this basis,the operating characteristics and microscopic features of buses stopping are discussed.Four bus stops with different forms and locations in Harbin are randomly selected.The data on bus stopping and the parameters of the environment of bus stops are collected by means of manual surveys and video cameras to analyze the stopping characteristics of buses.Secondly,a kinematic model is developed based on Ackermann Principle.The constraints on obstacle avoidance,kinematic constraints and start/end positions are discussed separately for the absence and presence of vehicles of bus stops.The geometric curves commonly used in trajectory planning are compared and analyze.On the basis of ensuring that the curvature of the trajectory can vary continuously,the quintic polynomial curve is chosen to develop the model of trajectory planning.Taking the vehicle stability and travel efficiency into account,the nonlinear multi-constraint equations for the trajectories of bus moving in and out of stops are designed with the shortest longitudinal displacement and minimum maximum curvature of the trajectory as the optimization objective.Again,the Genetic Algorithm is used to find the optimal solutions for the parameters of the trajectory profiles.The MATLAB software is applied to simulate the trajectories of bus moving in and out of stops under two conditions to verify the effectiveness of the method for trajectory planning.A follow-the-bus experiment is designed to collect data on the actual trajectories of buses moving in and out of stops and evaluate the effect of the trajectory planning with Mean-Absolute-Percentage Error(MAPE).Finally,the tracking controllers for bus moving in and out of stop are designed with the algorithms for low-speed condition named Model Predictive Control,Pure Pursuit and Stanley.Then,the simulation model is developed with the MATLAB/Simulink’s S-function.The tracking performance of three controllers are compared and analyzed to determine the method for tracking control of the trajectory of buses autonomously moving in and out of stops.