- 140 views
- 268 downloads
Multi-Agent Asynchronous Real-time Maze-solving
-
- Author / Creator
- Alian, Hamid
-
In recent years, mazes have been used to study robot behavior by assessing their ability to navigate a maze using various methods. The act of finding a path through a maze from beginning to end is known as maze solving. Some maze-solving methods are employed for use by an agent with no previous knowledge of the maze, while others are designed for use by someone or a computer program capable of seeing the whole maze at once. We consider a situation in which multiple agents are randomly distributed inside an arbitrary rectangular maze and have no previous knowledge of the maze. We provide a solution in the form of an algorithm for the agents to cooperate collaboratively to discover and achieve the hidden goal. We divide our algorithm into two major phases and present rules for each, with each agent is programmed to follow these rules individually. We explained the algorithm's implementation by addressing the challenges we have such that the agents can follow the algorithm in such a way that all agents may move simultaneously. We evaluate our approach using a computer simulation of a square-shaped maze with varying sizes and a variable number of agents. The algorithm performs well in the simulation, is efficient, and reflects the trade-off between utilizing a single agent and multiple agents. We provided the solution's results for both phases. We then validate our algorithm on a physical system consisting of a real maze and many robots. Our solution is primarily based on cooperating and working parallel with all agents.
-
- Subjects / Keywords
-
- Graduation date
- Fall 2022
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.