Seminar: Pattern Formation in Anonymous Swarms of Robots

Nicholi Shiell
M.Sc. Candidate
Supervisor: Dr. Andrew Vardy

Pattern Formation in Anonymous Swarms of Robots

Department of Computer Science
Monday, November 28, 2016, 11:00 a.m., Room EN 2022


Abstract

The ability to form patterns is useful to maximize the sensor coverage of a team of robots. Current pattern formation algorithms for multi-robot systems rely on the ability of robots to uniquely identify each other. This requirement increases the sensory and computational abilities of the individual robots, and reduces the scalability, robustness, and flexibility of the pattern formation algorithm. The research presented in this seminar focuses on the development of a novel pattern formation algorithm called the Dynamic Neighbour Selection Algorithm (DNSA). The DNSA does not require robots to be uniquely identified, thus improving its scalability, robustness, and flexibility of the technique. The DNSA was developed in simulation, and demonstrated on a team of vision enabled Bupimo robots. The Bupimo robots developed as part of the research will be used in other research projects in the BOTS laboratory, and for other educational purposes.