Path planning method based on reinforcement learning algorithm in dynamic environment

Abstract

The invention discloses a path planning method based on a reinforcement learning algorithm in a dynamic environment. The method comprises the following steps of carrying out the modeling of an operation environment through a grid method according to a dynamic known environment, and setting a starting point and a target point according to an operation task, constructing an improved Q-learning algorithm based on the time variable t, and endowing the intelligent agent with a stop action,learning a dynamic known environment based on an improved Q-learning algorithm according to the coordinates ofthe starting point and the target point, and outputting a planned path,exploring and learning a dynamic unknown environment state through the intelligent agent and obtaining an output optimal path. After training is completed, the speed is very high, which is attributed to learning attributes of a machine learning algorithm. In the aspect of effect, no matter what conditions the environment is, the algorithm cannot generate collision, and the application of the algorithm in path planning is very wide; and as a result, since the intelligent agent is endowed with a new stop action, a better pathcan be found in the planning.

Description

technical field [0001] The invention relates to a path planning method, in particular to a path planning method based on a reinforcement learning algorithm in a dynamic environment. Background technique [0002] Path planning problems generally deal with known and unknown environmental conditions. Many existing algorithms can solve its path planning problem, but for the path planning problem in a dynamic environment (moving obstacles), the algorithm cannot quickly solve the problem in this scenario, and the given planned route may collide. [0003] For example, the general heuristic algorithm combines the advantages of depth search and breadth search, so that the pathfinding algorithm can find the optimal solution with a high probability while ensuring the speed. But the general algorithm cannot deal with the problem of dynamic environment. And in a complex environment, the search efficiency of the heuristic algorithm will become lower. At the same time, for obstacle avoi...

AI Technical Summary

Problems solved by technology

But the general algorithm cannot deal with the problem of dynamic environment

And in a complex environment, the search efficiency of the heuristic algorithm will become lower

At the same time, for obstacle avoidance planning in unknown dynamic environments, such as D*, Lifelong A* and other algorithms are just a timely planning

Images