Research On Autonomous Lane Change Control Of Intelligent Vehicle

The study of autonomous vehicle lane change can reduce traffic accidents and road congestion caused by vehicles during lane change,and has important research significance.This paper focuses on three aspects of autonomous vehicle lane change decision making,trajectory planning and trajectory tracking.Firstly,the motorway is simplified into two lanes and the lane change scenario is divided into two typical lane change scenarios of straight lane change and curve change.The behaviour of lane change is analysed,and speed expectation and distance expectation are introduced representing the generation of lane change intentions.From the perspective of safety,the minimum safe distance model is developed for lane changing in straight and curved lanes respectively,making lane change more feasible and safe.The minimum safe distance models for lane change on straight and curved roads are simulated for both uniform speed and uniform deceleration lane change situations.The simulation results show that the minimum safe distance model on straight roads cannot replace the minimum safe distance model on curved roads,and decelerating lane change are relatively more feasible under the same conditions.Secondly,based on the vehicle state information at the initial and end moments of the lane change,a trajectory planning model for lane change on straight and curved roads is established using a quintuple polynomial function,and the feasibility of the model is verified by offline simulation.In view of the uncertainty of the ambient vehicles,the trajectory replanning decision mechanism is established based on the safety perspective,considering that the host vehicle may collide with the ambient vehicles during the lane change process.Finally,a trajectory planning controller is designed to select the optimal time to balance lane change efficiency,comfort and safety.Thirdly,in order to consider the controller accuracy and real-time,a linear model predictive trajectory tracking controller is established through a simplified vehicle dynamics model,combined with model predictive control ideas.Then the objective function is designed,control quantity constraints,input quantity constraints and output quantity constraints are added,and the objective function is transformed into a quadratic programming problem for solution.Finally,the performance of the trajectory tracking controller is tested by a joint simulation experiment.The simulation results show that the controller has a good trajectory tracking effect,and the reasonable selection of controller parameters helps to improve the controller performance.Finally,a joint simulation experiment is built in Carsim and MATLAB/Simulink,and three different working conditions are designed to verify the autonomous lane change algorithm designed in this paper.The results show that the vehicle can change lanes autonomously on both straight and curved roads,and that it can re-plan the lane change trajectory according to the uncertainty of the environment during the lane change process,thus ensuring the safety.