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

为了提高智能车辆避障的安全性与舒适性,提出了一种基于改进势场模型与隐马尔可夫模型的避障方法。首先建立了考虑车辆特性、道路环境与行驶状态等因素的改进势场模型,预测碰撞风险的动态变化;其次利用隐马尔可夫模型进行避障方式决策,并在改进的势场模型中融入隐马尔可夫决策层,完成自车避障路径规划;然后利用模型预测控制方法对规划的避障路径进行实时跟踪,并在控制器中加入松弛因子与约束条件防止出现无最优解;最后应用联合仿真对提出的避障方法进行验证,结果表明,所提出的方法可在复杂交通环境下获得无碰撞的避障路径,实现动态避障的同时,提高了车辆的安全性与舒适性。

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 constructed 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 predictive 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.