Application of Q-learning in Autonomous Robot Obstacle Avoidance

DOI

10.2991/978-94-6463-823-3_36How to use a DOI?

Keywords

Q-learning; Autonomous robot; Obstacle avoidance

Abstract

In this paper, 8 different applications of QL in autonomous robot obstacle avoidance are introduced. A method that connects calculation of Q with state changing reduces the episodes spent on maze solving. Adding prioritized weight to samples can make the improved QL find the shortest path in a maze compared to original QL, A*, probabilistic roadmap, rapid exploring random trees and reinforcement learning. Double Q-learning avoids overestimation efficiently and converges quicker than original Q-learning. QL with dynamic rewards and QL with shortest distance prioritization decompose the reward into 2 parts and reduces the steps cost and convergence time. Deep Q-network (DQN) with prioritized weight improved the success rate of obstacle avoidance compared to original DQN and Q-learning. Combining the max operation or double estimator operation is possible with the stochastic double deep Q-network to achieve a faster convergence speed. Double delayed Q-learning demonstrates faster convergence and better estimation than double Q-learning and QL.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhitao Su
PY  - 2025
DA  - 2025/08/31
TI  - Application of Q-learning in Autonomous Robot Obstacle Avoidance
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 366
EP  - 375
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_36
DO  - 10.2991/978-94-6463-823-3_36
ID  - Su2025
ER  -