智能系统学报2023，Vol.18Issue(6)：1275-1286,12.DOI:10.11992/tis.202210033
复杂交通环境下智能车辆避障方法研究
Obstacle avoidance method for intelligent vehicles in complex environments
摘要
Abstract
An obstacle avoidance method based on an improved potential field model and hidden Markov model(HMM)is proposed to enhance the safety and comfort of intelligent vehicles during obstacle avoidance.First,the method con-structed an improved field model by considering typical elements of vehicle attributes,road environment,and driving states to predict the dynamic changes in collision risk.Further,HMM was used for obstacle avoidance decision making,and path planning was performed by incorporating HMM into the improved potential field model.Next,the model pre-dictive control was used to track the planning paths in real time,and a relaxation factor and constraints were added to the controller to prevent the problem of unavailable optimal solutions.Finally,the cosimulations by CarSim and MATLAB/Simulink were conducted to verify the effectiveness of the proposed method.Results show that the proposed method can enhance driving safety and ride comfort while obtaining collision-free driving paths in various situations and realizing dynamic obstacle avoidance.关键词
智能车辆/势场模型/隐马尔可夫模型/动态避障/路径规划/跟踪控制/模型预测/复杂环境Key words
intelligent vehicle/potential model/HMM/dynamic obstacle avoidance/path planning/tracking control/MPC/complex environments分类
信息技术与安全科学引用本文复制引用
李海青,李永福,郑太雄,李洪丞..复杂交通环境下智能车辆避障方法研究[J].智能系统学报,2023,18(6):1275-1286,12.基金项目
国家自然科学基金项目(62273067,U1964202) （62273067,U1964202）
国家重点研发计划项目(2018YFB1600500) （2018YFB1600500）
重庆市自然科学基金面上项目(cstc2020jcyj-msxmX0915) （cstc2020jcyj-msxmX0915）
重庆市教委青年项目(KJQN202100644). （KJQN202100644）
