Fuzzy and Neuro-fuzzy Based Co-operative Mobile robots
To listen to the presentation click play.
To view the presentation on full screen, position the mouse on the presentation, right-click on it,
select zoom and then full screen.
Author : D T Pham, M H Awadalla and E E Eldukhri
This paper focuses on the development of intelligent multi-agent robot teams that are capable of acting autonomously and of collaborating in a dynamic environment to achieve team objectives. A biologically-inspired collective behaviour for a team of co-operating robots is proposed. A modification of the subsumption architecture is proposed for implementing the control of the individual robots. The paper also proposes a fuzzy logic technique to enable the resolution of conflicts between contradictory behaviours within each robot. Furthermore, the paper proposes a neuro-fuzzy based adaptive action selection architecture that enables team of robot agents to achieve adaptive cooperative control to perform two proof-of-concept co-operative tasks: dynamic target tracking and box-pushing. Simulated and real experiments have been conducted to validate the proposed techniques.
| Attachment | Size |
|---|---|
| PID154784.pdf | 196.22 KB |
login or register to download the paper.
You said that" A multi-channel infrared communication system to exchange messages among mobile robots is developed"
Infra-red communication needs straight line between transmitter and reciever without obstacles how did you get rid of this problem? and Why didn,t you use radio communication instead of infra-red?
Please edit your entry again and check the "List" checkboxes for both the PDF and the PPT attachments at the attachments area. It seems there is a bug? that deselects the checkbox for the PDF when you submit a new item and you have to select them again manually and press Submit. If you accidentaly selected the checkbox that removes the items and removed the PDF you can either resubmit it from the submission website (hope it's still alive) or from the confirmation e-mail you had been sent by the submission website, or ask the IPROMS people to reupload the PDF
Dear Afshin,
In the first experiment, two robots have to co-operate to push a box across the room with no obstacles in the environment. In the second experiment, one of the robots is obstructed to investigate how the other one dynamically re-selects its acions in response to changes in the mission situation. In both experiments, the Webots Simulation Tool was used to implement this task. After the robots locate the box, they were instructed to move left and/or right until they positioned themselves within reasonable distances from the box ends. During simulation, it is assume that the robot at the left end of the box will start to push first.
Infra-red sensors and cheaper and simple to implement. It's true that they have limiitations, however, in this work they serve the purpose of attaining cooperation and coordination of robots actions. In future work, more sophisticated communication and control infrastructures would be sought.
hi:
Is these robots self relying or is there a master/slave program gouvern them?
best regards
yi
Hi
Thanks for the good paper, I hope we could see the presentation very soon too.
I have a question abd localisation of robots around the object (box). How they localise themselves. Is it just go to the end of it or ... .
Many Thanks,
Afshin