Loading...

Passage-Way Identification and Crossing Control of an Experimental Swarm Robot

Balaghi Enalou, Mohammad Hadi | 2015

780 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47696 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Alasti, Aria; Vatankhah, Ramin
  7. Abstract:
  8. The aim of the proposed thesis is to devise a method of passage-way identification and crossing control of an experimental swarm robot. In the previous works done in the Sharif University of Technology, The main focus was on the collective behaviours than can be emeged from a swarm robot. Collective behaviours are such a primitive behaviours that are combined in order to lead to a real mission. In this investigation, the main focus is on the design of a mission. In this way, by combining the aggregation-divegance, chain formation, leader following and wall following collective behaviours, the desired is fulfilled. The algorithms is so that in the first stage, the agents of the swarm diverged in the environment. This meanse that the robots obey the aggregation-divergance algorithm in this stage. If an agent arrives to the environment boundary which is distinguished with a wall it will start the wall following algorithm. The wall following algorithm makes the robots to maintain at an equilibrium distance from the wall and walking against the wall, simultaneously. The first robot arrives to the passage-way, informs the other robots and contributes in the chain formation with the other agents. In this chain, the agent finds the passage-way, will stand at the head of the chain and acts as the leader. When the chain is formed, the leader moves toward the crossing-way and the other tries to mainmtain the chain, so they are force to follow the leader. Each agent of the swarm arrives to the beginning of the crossing-way, starts the wall following algorithm to go out of the environment through the crossing-way. The resuls of the simulation, demonstrate the perfect performance of the proposed algorithm. In order to implement this algorithm in the practice, the swarm robot of the control lab in the mechanical department of Sharif University of Techanology is developed. In this way, by the aid of central image processing the positions and directions of the robots are calculated. Then these information is sent to the whole robots. Finally, Aggregation-divergance, chain formation, leader following, wall following and identification of passage-way and crossing control algorithms are implemented on this system. The results shows satisfactory behavior of the swarm
  9. Keywords:
  10. Image Processing ; Swarm Control ; Passageways ; Agregation-Divergence Algorithm ; Wall Following ; Passageway Identification ; Passageway Crossing Control

 Digital Object List

 Bookmark

No TOC