Securing emergent behaviour in swarm robotics

Swarm robotics is the study of how a large number of relatively simple robots can be designed so that a desired collective behaviour emerges from the local interactions among robots and between the robots and their environment.

Swarm emergent behaviour is one of the aspects that most distinguishes swarm robotics from ad hoc networks and poses many challenges on swarm security. By modelling a swarm using random graphs, and ensuring that events are recorded securely in a hash chain, we can allow robots to identify “bad” robots with a high probability, while ensuring that “good” robots were not adversely affected. This goes some way towards protecting emergent behaviour, by limiting the influence of bad robots locally.

In the scenario where robots may have uncorrelated interactions with each other binomial random graphs provide a reasonable model, and we can use a series of such graphs over time to model movements. However, we would like to examine more complex situations, for example, where there are larger and less homogeneous networks, or where some robots have special roles, or where bad robots are modified to have additional capability - what models might be suitable then?

A background in discrete mathematics and some knowledge of cryptography and programming would be useful.

Please contact Dr Siaw Lyn for further information