Obstacle Avoidance Trajectory Planning And Tracking Control For Intelligent Vehicle

With the continuous improvement and iteration of intelligent vehicles,its vast prospects for applications in the automotive industry has caused a huge wave.Improving driving safety has always been the primary goal pursued by intelligent driving vehicles,and obstacle avoidance function is an essential guarantee for achieving this goal.When an intelligent driving vehicle detects obstacles,the trajectory planning module dynamically plans a reasonable avoidance trajectory based on the current vehicle motion state and instructions from the decision-making module,while the trajectory tracking module calculates the corresponding control quantity based on the planned trajectory to complete the trajectory tracking.As indispensable modules for achieving obstacle avoidance function,the trajectory planning and trajectory tracking modules are studied involved in obstacle avoidance problems,the performance of the algorithm is verified and analyzed them by a simulation platform.Firstly,the existing curve fitting methods is compared in trajectory planning,the Sigmoid curve is chose as fitting method for path planning.A vehicle kinematic model and a threedegree-of-freedom dynamic model are established,which are respectively incorporated into the trajectory planning and trajectory tracking modules,and serve as the basis for the design of the trajectory planning and trajectory tracking modules.Secondly,based on traditional artificial potential field,a road potential field composed of road boundary potential field,road branch potential field,main lane attraction potential field,and obstacle repulsion potential field is established.By incorporating the Sigmoid curve fitting,kinematic model,and gradient descent method into the road potential field,the trajectory planning module is developed,and the simulation results show that the trajectory planning module can reasonably plan obstacle avoidance trajectories under the constraints of road structure and vehicle kinematics.Thirdly,applying the Nonlinear Model Predictive Control theory to the construction of the trajectory tracking controller,and considering both the tracking accuracy and calculate time consumption.of the controller,the multiple shooting method is used to transform the trajectory tracking problem into a Nonlinear Programming Problem under multi-objective constraint conditions.By setting with some simulation scenarios and comparing the tracking effect of the Linear Predictive Control model,it shows that the established Nonlinear Model Predictive controller not only improves the tracking accuracy but also satisfies the requirement of calculate time consumption.Finally,the simulation results show that a reasonable obstacle avoidance trajectory can be planned in a variety of obstacle avoidance scenarios considering vehicle kinematic constraints and road structure constraints,At the same time,the trajectory tracking module can effectively track the obstacle avoidance trajectory.