A Path Planning Method for Mobile Robots Combining Deep Autoencoder and Q-Learning Algorithm

Abstract

A path planning method for a mobile robot combining a deep autoencoder and a Q-learning algorithm, the method includes a deep autoencoder part, a BP neural network part, and a reinforcement learning part. The deep autoencoder part mainly uses the deep autoencoder to process the image of the robot's environment, obtain the characteristics of the image data, and lay the foundation for the subsequent realization of the cognition of the environment. The BP neural network part mainly realizes the fitting of reward value and image feature data, and realizes the combination of deep autoencoder and reinforcement learning. The Q-learning algorithm learns through interaction with the environment, acquires knowledge in an action-evaluation environment, and improves the action plan to adapt to the environment to achieve the desired purpose. The robot realizes autonomous learning through interaction with the environment, and finally finds a feasible path from the starting point to the end point. The invention improves the ability of the system to process images, and realizes the cognition of the environment through the combination of the deep automatic encoder and the BP neural network.

Description

technical field [0001] The invention relates to a path planning method for a mobile robot combining a deep autoencoder and a Q learning algorithm, and belongs to the field of path planning for robots. Background technique [0002] Path planning is a fundamental problem in the field of mobile robotics. Mobile robot path planning refers to how to find an appropriate motion path from a given starting point to an end point in a working environment with obstacles, so that the robot can bypass all obstacles safely and without collision during the motion process. [0003] With the development of robot technology, robots have begun to be applied to unknown environments. Compared with the research on path planning of mobile robots in known environments, the exploration of unknown environments has brought new challenges. Since the robot does not have prior knowledge of the environment in an unknown environment, the mobile robot will inevitably encounter various obstacles during the p...

AI Technical Summary

Problems solved by technology

All of the above methods require manual image processing

The traditional manual extraction of image features is a very laborious and heuristic (requires professional knowledge) method. Whether or not good features can be selected depends largely on experience and luck, and its adjustment takes a lot of time

Images