Reinforced learning path planning algorithm based on potential field

Abstract

The invention, which belongs to the field of intelligent algorithm optimization, provides a reinforced learning robot path planning algorithm based on a potential field in a complex environment, thereby realizing robot path planning in a complex dynamic environment under the environmental condition that a large number of movable obstacles exist in a scene. The method comprises the following steps:modeling environment space by utilizing the traditional artificial potential field method; defining a state function, a reward function and an action function in a Markov decision-making process according to a potential field model, and training the state function, the reward function and the action function in a simulation environment by utilizing a reinforcement learning algorithm of a depth deterministic strategy gradient; and thus enabling a robot to have the decision-making capability of performing collision-free path planning in a complex obstacle environment. Experimental results showthat the method has advantages of short decision-making time, low system resource occupation, and certain robustness; and robot path planning under complex environmental conditions can be realized.

Description

technical field [0001] The invention belongs to the field of intelligent algorithm optimization and relates to a potential-field-based reinforcement learning path planning algorithm for complex dynamic environments. Background technique [0002] The path planning method refers to the method for the robot to plan an optimal path from the starting point to the target point under the premise of no obstacle collision. Path planning is an optimization problem that satisfies constraints. The optimization indicators usually include the shortest time, the best route, and the lowest energy consumption. The algorithm needs to have certain characteristics such as complexity, randomness, and multiple constraints. According to the algorithm model of path planning, it can be divided into traditional methods and intelligent methods. Commonly used traditional methods include grid method, artificial potential field method and topological space method, etc. Commonly used intelligent methods i...

AI Technical Summary

Problems solved by technology

This method is simple and practical, but it is computationally intensive, so it is only suitable for simple environments

The quadtree method improves the grid map method and divides the environment map into four parts. The quadtree data structure enables the environment to be quickly modeled. However, when many small obstacles exist in the environment, the tree structure will deepen and the amount of calculation will be increased. increase accordingly

[0008] After the research on the existing robot path planning methods, it is found that the current path planning methods still have certain limitations in the face of complex dynamic environments.

When the magnitude of obstacles reaches a certain level, the calculation amount of the algorithm increases, the calculation efficiency decreases, and it is difficult to plan a better path

Images