基于Frenet坐标系的智能车运动规划研究

Motion Planning of Autonomous Vehicle Based on Frenet Coordinate System

自动驾驶技术作为车辆智能的核心,在车辆工程领域掀起研究高潮。运动规划作为决策模块的重要组成部分,负责生成车辆的局部运动轨迹,是决定车辆行驶质量的直接因素。将Frenet坐标系用于智能车运动规划问题研究,用五次多项式建立智能车运动规划模型,在自主变道场景,通过采样智能车始末状态,得到可行轨迹集合,建立质量评估函数从中筛选出最优轨迹,开发和实现一种高性能、低开销的运动规划算法。

As the core of vehicle intelligence,autonomous driving technology has raised a research climax in the field of vehicle engineering.As an important part of the decision-making module,motion planning is responsible for generating the local motion trajectory of vehicles,which is a direct factor to determine the driving quality of vehicles.The Frenet frame used in the study of intelligent vehicle motion planning problem,with five times polynomial intelligent vehicle motion planning model is set up,in the independent variable scenario,by sampling the beginning and end of the smart car state,get the feasible path set,set up quality evaluation function through the optimal trajectory,to develop and implement a motion planning algorithm of high performance and low cost.