高速公路弯道换道决策及运动规划优化

弯道换道决策及运动规划算法主要影响自动驾驶汽车的安全性和操纵稳定性。针对高速公路弯道换道场景决策的安全性和行驶效率不够高的问题，提出新的基于主车相对前车的驾驶不满意度的决策算法。为了提高运动规划算法实时性，采用路径-速度解耦框架进行主车换道轨迹规划。对于路径规划，选择五次多项式曲线，采用考虑安全、舒适和高效性的4个换道路径评价指标，实现最优路径规划。对于速度规划，结合动态规划与二次规划优化获取平滑速度规划曲线。仿真结果表明基于驾驶不满意度的换道决策模型能选择更高效和安全的行驶方式。在典型的主车换道场景，主车最大质心侧偏角，最大横摆角速度的数值均小，表明换道轨迹规划算法能确保主车换道的安全性和操纵稳定性。

Decision making and optimization of trajectory planning of lane change on highway curve

Decision making and motion planning algorithms play a crucial role in determining the safety and handling stability of autonomous vehicles, particularly on highway curves. Addressing safety and driving efficiency concerns in the decision-making process for highway lane changes, this study proposes a driving dissatisfaction decision algorithm based on the relative driving dissatisfaction of the ego-vehicle compared to the preceding vehicle. To improve the real-time performance of the planning algorithm, a path-speed decoupling framework is adopted for lane change trajectory planning. The path planning utilizes the quintic polynomial curve, incorporating four path evaluation indicators that considered safety, comfort and efficiency to achieve optimal path planning. Speed planning involves obtaining a smooth speed curve through a combination of dynamic programming and quadratic programming optimization. Simulation results show that the lane change decision model based on driving dissatisfaction can choose a more efficient and safer driving mode. In typical lane changing scenarios, both the maximum centroid sideslip angle and maximum yaw rate of the ego-vehicle are small, indicating that the lane change trajectory planning algorithm can ensure the safety and handling stability of the ego-vehicle during the lane change process.