Title
The use of GA for parameters generation in fault-tolerant algorithm for manipulator control in case of axis failure
Authors
[ 1 ] Instytut Technologii Mechanicznej, Wydział Inżynierii Mechanicznej, Politechnika Poznańska | [ P ] employee
Scientific discipline (Law 2.0)
[2.9] Mechanical engineering[2.10] Environmental engineering, mining and energy
Year of publication
2025
Published in
Article type
scientific article
Publication language
english
Keywords
EN
- Fault tolerant control (FTC)
- Robots control
- Axis failure
- genetic algorithm (GA)
Abstract
EN This paper presents a new fault tolerant control algorithm for the generation of a robot motion trajectory in an environment with obstacles, when one or two robot axes fail. The proposed method allows the robot to be controlled with obstacle avoidance in the event of axis failure at any time during the motion. An additional requirement for the algorithm was to determine, from among the generated steps that could be executed at a given point, the shortest path of the robot to the final position. However, due to the limitations of the GA used in the learning process, the total robot path was not the globally shortest. The algorithm works step by step. It uses a model of the robot working in the environment with obstacles and the reward function, defined in this article, to select the best possible robot move for each step. Two parameters are applied in this function, which values are generated using a genetic algorithm. At first, the proposed control method has been tested in the simulation using six arm robot model. The results are validated on a real Mitsubishi robot.
Date of online publication
11.05.2025
Pages (from - to)
105062-1 - 105062-12
Comments
Article Number: 105062
Ministry points / journal
140
Impact Factor
4,3 [List 2023]
System created by Poznań University of Technology
and Poznan Supercomputing and Networking Center
Log in through eKonto to add to SIS