Research On Trajectory Planning And Control Of Intelligent Vehicles On Urban Structured Roads

Under the blessing of new sci-tech revolution and industrial transformation,intelligent vehicles are in the stage of rapid technological evolution and accelerated industrialization.As the key technology of intelligent vehicles,trajectory planning and tracking control technology directly determine whether intelligent vehicles can drive safely and comfortably on the road.This paper takes the urban structured road as the research scene,and carries out the following research:Research on vehicle position estimation algorithm: A vehicle position estimation model based on Kalman filter is established.According to the curvature,the vehicle kinematics model with constant acceleration and the vehicle kinematics model with constant turn rate and acceleration are respectively used as the system model of Kalman filter,which aims to improve estimation accuracy.Research on trajectory planning algorithm: To facilitate understanding of urban structured road scenarios and simultaneously reduce planning difficulty,the trajectory planning is decoupled into two sub-problems,path planning and speed planning.For the path planning,the obstacle aggregation model is firstly proposed to improve the rationality of decision-making and the stability of the path.Then,the sampling space is constructed in the SL space,and the cost function is designed to evaluate the path points to obtain a reference path.The multi-objective optimization problem of path optimization takes the sampling space and differential constraints as constraints,and finally solves the optimization problem by quadratic programming.For speed planning,the reference speed along the desired path is firstly planned in the Cartesian coordinate system.In order to ensure that the reference speed can match the desired path and has high traffic efficiency,the planning process only considers the acceleration,road speed limit and curvature constraints,then a multi-objective optimization problem of speed optimization is constructed,and constraints include vehicle kinematics constraints,obstacle avoidance constraints considering the uncertainty of the longitudinal motion of obstacles,and the reference speed constraints.The optimization problem is solved by quadratic programming finally.In this paper,the curvature is used as a constraint in the form of a reference speed,so that the velocity programming can deal with nonlinear constraints within the framework of linear optimization,which has certain engineering significance.Research on trajectory tracking control algorithm: Based on the decoupling of dynamics of vehicles,a path tracking controller and a velocity tracking controller are designed respectively in paper.For path tracking control,a feedforward controller based on previewfollowing theory and an error-based feedback controller are designed in this paper.Based on the human-vehicle-road closed-loop system,the feedforward controller introduces adaptive correction method of preview time.The feedback controller takes the heading angle error as input to compensate steering wheel angle.For speed tracking control,a speed tracking controller based on hierarchical control is proposed.The decision layer calculates the desired acceleration based on model predictive control,and the execution layer conducts control mode arbitration by introducing the vehicle coasting curve to make the vehicle run smoothly.In the execution layer,the driving module parses the desired acceleration into the desired motor torque according to the longitudinal dynamics model on full work conditions,and the braking module directly requests the desired deceleration for actuator.Simulation and verification of trajectory planning and tracking control algorithms:A cosimulation platform based on SCANe R-ROS-Linux is builded,and simulation tests on the trajectory planning and tracking control algorithms are conducted with typical working conditions of urban structured roads.According to the simulation results,the trajectory planning algorithm proposed by this paper can provide a reasonable desired trajectory,and the designed trajectory tracking controller can also stably track the desired trajectory.According to the simulation results,the algorithms proposed by this paper can meet the needs of trajectory planning and tracking control in urban structured road scenarios.