Bp 362 en
Group intelligence algorithms for use in multi-robot tasks
Author: Vojtěch Pavlík
This thesis covers modifications of the Particle Swarm Optimization (PSO) algorithm, which allows its direct use to control swarms of real robots. In the presented approach, individual robots are considered as particles in the PSO swarm. Contrary to a virtual swarm of dimensionless entities in the classical PSO, in engineered PSO the constraints of the real robotic swarm need to be considered and implemented. Specifically, the motion model of real robots, robot movement restrictions, restrictions given by relative locations of the robots, communication between neighbours in a swarm, the size constraints of the robots and their interaction have to be considered. The next step is the modification of the algorithm for the needs of the robotic platform SyRoTek.
- Pavlík Vojtěch, tel: +420 721 420 119 ,mailto:pavlivo1@fel.cvut.cz
