【作者】 赵园； 晏雪刚； 邹阳； 文世喜； 张远鹏； 赵辉；

【Author】 ZHAO Yuan;YAN Xuegang;ZOU Yang;WEN Shixi;ZHANG Yuanpeng;ZHAO Hui;School of Information and Engineering, Dalian University;College of Computer and Information Engineering, Tianjin Normal University;School of Control Engineering, Northeastern University at Qinhuangdao;School of Artificial Intelligence, Shenyang University of Technology;

【机构】 大连大学信息工程学院； 天津师范大学计算机与信息工程学院； 东北大学秦皇岛分校控制工程学院； 沈阳工业大学人工智能学院；

【摘要】 本文考虑了高速公路上自动驾驶车辆针对静态障碍物避碰问题,提出了一种两阶段规划-控制方法,该方法首先对车辆的速度控制和转向控制进行解耦,在第一阶段进行速度规划,然后在第二阶段通过模型预测控制(Model Predictive Control,MPC)跟踪采用余弦方法规划的避碰轨迹,并通过坐标变换将该方法从直线道路扩展到了曲线道路上。本文最后设计了两组对比实验,仿真结果验证了所提方法的有效性和稳定性。更多还原

【Abstract】 This paper investigates the problem of static obstacle avoidance for autonomous vehicles on highways and proposes a two-stage planning-control method. In this method, the control of vehicle speed and steering is decoupled. In the first stage, speed planning is performed, and in the second stage, a model predictive control(MPC) is employed to track the planned avoidance trajectory. The designed avoidance trajectory is in the form of a cosine function and is extended from straight roads to curved roads through coordinate transformation. Two sets of simulations are conducted in this study, and the simulation results confirm the effectiveness and stability of the proposed method.更多还原