Abstract
Robot swarms, consisting of large numbers of individual robots that collectively work towards a common goal, must be autonomous in order to carry out their task without regular human input. Further, there is a requirement that such swarms be autonomic, capable of self-management to enable operation in distant, complex or changing environments.Underpinning the ability of a swarm to cooperate and adapt is the communication between individual robots, enabling information gathered at a local level to be shared with the swarm as a whole, and for swarm-level perspectives to be available. Wireless communication between robots that relies upon a fixed transmission range may be subject to connectivity loss, or place restraints on robot motion that may negatively affect the swarm’s performance. If the range is too large, it may be wasteful and cause interference. Robot loss and unknown environments may mean transmission ranges that were once ideal are no longer the case.
The contribution of this thesis is the development of a method for achieving a dynamic transmission range for data transfer in a decentralized robot swarm, based on the existing concept of autonomic pulse monitoring. Through simulation, the problem of a fixed transmission range is demonstrated, and then the Autonomic Pulse Communications (APC) system is tested in three scenarios: foraging robots; subswarm formation; and a simulation of the NASA Prospecting Asteroid Mission concept. The results show that the APC system can select a transmission range for each robot using only information gathered from nearby robots, and use this to transfer data throughout the swarm.
Future research in this area may improve the efficiency of the data transfer behaviour by adopting selective transmission techniques, while experiments using real robots will be required to bridge the reality gap.
Date of Award | Jul 2023 |
---|---|
Original language | English |
Sponsors | Department for the Economy |
Supervisor | Glenn Hawe (Supervisor), Roy Sterritt (Supervisor) & George Wilkie (Supervisor) |
Keywords
- Autonomic computing
- Swarm robotics
- Self-adaptation