Please use this identifier to cite or link to this item:
Title: Design and Implementation of Efficient Coordination Algorithms for Multi Robot Systems
Researcher: Gautam, Avinash
Guide(s): Mohan, Sudeept
Keywords: Computer Science, Coordination Algorithms, Multi-Robot Systems
University: Birla Institute of Technology and Science
Completed Date: 2016
Abstract: In multi-robot systems, individual robots cooperate with each other and work as a team to solve complex problems that are otherwise difficult to solve for a single mobile robot. There are several motivations for developing multi-robot solutions. Some of them are: newline1. Some tasks are complex in nature and therefore it is difficult for a single robot to accomplish them, for example, parallel and simultaneous transportation of load. newline2. Some tasks are monotonous, inherently parallel, and hazardous, for example, floor cleaning, lawn mowing, field harvesting, patrolling and battle field surveillance. newline3. It is easier to build many simple robots rather than building a single monolithic omnipotent robot. newline4. The use of multiple robots increases the robustness of the system i.e., makes the system more fault tolerant. newlineIn this thesis we have proposed and implemented coordination algorithms for multi-robot systems for solving problem like, geometric pattern formation, online terrain coverage and load balanced task decomposition and allocation. We have shown that a properly coordinated robot team achieves better performance by efficiently utilizing the available system resources i.e. the robots. newlineThe first problem investigated in this thesis is that of geometric pattern formation using multiple mobile robots. Specifically, the uniform circle formation problem has received considerable attention. Many researchers have addressed this problem from computational perspective and have suggested algorithms while treating the robots as abstract computational entities. We have designed and implemented a decentralized algorithm, referred to as STATE, for uniform circle formation using multiple mobile robots. One of the benchmark algorithm proposed by Défago and Konagaya is re-implemented on our multi-robot test-bed. The STATE algorithm is shown to perform better than the Défago and Konagaya s algorithm. The better performance of the STATE algorithm is due to its order preserving scheduling policy for multi-robot synchronization.
Pagination: 198p.
Appears in Departments:Computer Science & Information Systems

Files in This Item:
File Description SizeFormat 
final_thesis.pdfAttached File4.65 MBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.