基于B样条曲线的智能汽车时空联合规划方法

打开文本图片集

中图分类号：U471.15 文献标志码：A

Abstract： To address the problems of inefficient curvature optimization and the challenge of balancing obstacle avoidance accuracy with kinematic feasibility in complex environments，a Spatio-Temporal B-spline Planning （ST-BSP） algorithm based on B-spline curves was proposed. By introducing a temporal mapping to the control point sequence，the algorithm achieved joint optimization in both spatial and temporal dimensions. The trajectory optimization process was divided into two phases. In the first phase，under the safe driving corridor framework，an adaptive iterative strategy was designed by integrating the obstacle avoidance cost and the second-order discrete curvature. This approach effectively handled diverse obstacle scenarios， generated smooth trajectories with continuous and bounded curvature. In the second phase，a time redistribution strategy was applied to detect and correct violations of kinematic constraints，thereby enhancing trajectory feasibility. Simulation results demonstrate that， compared with existing decoupled optimization-based methods， the proposed ST-BSP algorithm achieves superior trajectory rationality， smoothness，and obstacle avoidance performance in complex scenarios such as continuous obstacle bypassing and narrow passage traversal. Additionally，it reduces computation time by 78.4% ，indicating strong potential for application in unstructured and complex road environments.

Keywords： intelligent vehicles;trajectory planning;B-spline curve;safe driving corridor;iterative solving strategy

随着科学技术的发展，自动驾驶逐渐参与到人们日常通勤并且降低了交通事故的概率，成为近年来的热门研究课题。（剩余8554字）