Our group primarily works with rovers (ground-based robots), and we have obtained several ‘Leo Rovers’ for our experiments. Previously we have also worked with ‘Kilobots’ for conceptual swarm research. As our research develops, we will examine heterogeneous swarms (i.e. different robot types) including e.g. quadruped (4-legged) robots and drones in our swarm deployments.
Effective Patrolling Behaviours
We want to deploy teams of robots into real-world environments to monitor activity to ensure anything anomalous can be detected and investigated. In a multi-robot / swarm context, there is a clear opportunity to exploit cooperation between robots to verify anomalous findings by sending follow-up patrols.
The question of what patrolling behaviours best enable this remains quite open, especially when robots can have heterogeneous paths, and are operating in an adversarial context.
We investigate patrolling behaviours in simulation (e.g. Gazebo) before testing controllers on real robots.
Managing the indoor-outdoor transition
Patrolling robots can work in both indoor and outdoor contexts, switching between them as they move in and around complex building sites. However, it is an engineering challenge to obtain effective SLAM (simultaneous localisation and mapping) when switching between indoors and outdoors – e.g. lighting conditions may drastically change, GNSS will be available/denied, and so forth.
There are also opportunities to improve SLAM using the presence of multiple robots which can share their information. We are focused on real-world swarm engineering that can operate both indoors and outdoors.
Evolving behavioural plasticity
Based on our understanding of behavioural biology, some animals demonstrate phenotypic plasticity (flexibility) in their behavioural types, for example to be ‘bold’ or ‘shy’ in how they respond to risk.
We are investigating whether and how behavioural plasticity in these fundamental ‘personality types’ can emerge in evolutionary swarm robotics. We are evolving controllers in simulated environments (initially using the ‘RoboRobo’ simulator of Bredeche et al., pictured), and then plan to demonstrate the relevance of behavioural plasticity in real robots. These robots may be Leo Rovers (pictured above) or more typical lab-based swarm robots.