Obstacle Bots for Swarm Robots


In swarm robotics the major barrier is that the researchers have to do a lot of hardware implementation prior to their projects. This particular project we are going to come up with obstacle bots for a swarm robotic arena which is a part of a swarm robots platform project which eventually solves the above mentioned problem.

The main goals of this project are to monitor automated obstacle bots and move them to the desired positions with a user-friendly interface by avoiding collisions.

Robots are localized with an overhead camera setup in order to create very challenging patterns using movements in a self-created potential surface with collision-avoiding algorithms on optimized stochastic gradient descent and particle repulsion theory. Also these obstacles can be programmed to be static or dynamic. Dynamic obstacles can model scenarios that have a motion in the obstacle.

This project was funded by the Prof. Suhada Jayasuriya Project Support Fund through PEFAA https://pefaa.net/.