Abstract—Robots are used to move stored items in Automated Storage (AS) to shelve and pickup items in warehouse depicted shelves. In this situation and many others, it is important to follow the shortest way so that the smallest time possible is taken to achieve the task. In this research, a Fuzzy Logic system with Q-Learning Reinforcement in order to achieve a better overall system. While Fuzzy Logic can be used alone for robot navigation, Reinforcement learning helps to adjust the fuzzy rules and refine them towards two main purposes: reach the final goal, while avoiding difficult obstacles such as traps. This is done as an enhancement on our previous work where Fuzzy Logic system was used alone. Simulation results are added to support the work done. It proved that this new system is much better than the previous one. Highlighting key parameters or features of simulation results show that the system achieved 33% more optimized time in addition to avoid stalled/unsuccessful navigation in some difficult situations, thus demonstrating the system’s success.

Keywords—fuzzy logic, fuzzy reinforcement learning, path planning, Q-Learning, robotics

Cite: Chadi F. Riman and Pierre E. Abi-Char, "Novel Fuzzy Reinforcement Algorithm for Mobile Robot Navigation in Automated Storage," International Journal of Mechanical Engineering and Robotics Research, Vol. 13, No. 2, pp. 284-295, 2024.

Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.