Method for planning and tracking local obstacle avoidance path of autonomous vehicle

Abstract

The invention relates to a method for planning and tracking a local obstacle avoidance path of an autonomous vehicle. The method comprises the following steps: acquiring a global reference path; planning a local path of the autonomous vehicle through an RRT local path planner; and stably and effectively tracking an ideal local obstacle avoidance path by adopting a personified T-S fuzzy path tracking strategy. According to the invention, a vehicle dynamics model is fused into an RRT algorithm, and the personified driver behavior dynamics model and the fuzzy algorithm are fused in path tracking control so that the autonomous vehicle can effectively track an ideal obstacle avoidance path in a high-dynamic road environment, and meanwhile, the dynamics stability of the vehicle can be well ensured; and a theoretical basis preparation is made for path planning and tracking of the autonomous vehicle applied to urban road vehicle driving with dense pedestrians, field environment engineering vehicle obstacle avoidance driving and even battlefield military vehicle actual obstacle avoidance driving for avoiding enemy firepower in the future.

Description

technical field [0001] The invention relates to the traveling technology of an automatic driving vehicle, in particular to a planning and tracking method for a local obstacle avoidance path of the automatic driving vehicle. [0002] technical background [0003] The unmanned driving technology of self-driving vehicles, especially path planning and path control, is the foundation and support of the development of intelligent networked vehicles, and it is necessary to focus on key research and breakthroughs at this stage. Today's domestic and foreign researchers have conducted relatively in-depth research on path planning and path tracking in static environments, but path planning and tracking algorithms for local obstacle avoidance in complex dynamic environments are rarely involved. [0004] Path planning methods for autonomous vehicles are generally divided into global path planning and local path planning. In complex dynamic environments, local path planning should be selec...

AI Technical Summary

Problems solved by technology

As stated in the Chinese patent 201910952980.8 "Mixed Path Planning Algorithm for Unmanned Vehicles", this path planning method based on random search ideas can well solve the problem of local minima, but this method lacks the real-time state feedback of the vehicle itself. In actual application, it is very likely that the planned ideal path is difficult to be tracked by the path controller or the tracking effect is not good

Furthermore, this method is time-consuming to compare and analyze one by one, consumes a lot of computing resources, and the real-time performance cannot be effectively guaranteed, especially in the case of obstacle avoidance in a dynamic environment, the planned path cannot be updated in time, causing the vehicle to collide with moving obstacles

Images