SwaCIL

Swarm & Computational Intelligence

Laboratory

Swarm & Computational Intelligence Lab (SwaCIL) is a member of the University of Manchester Robotics Group based in the Dept of Electrical & Electronics Engineering. Its main focus is to investigate bio-inspired swarm intelligence mechanisms for use in real-world applications. SwaCIL formed in 2018 and the target applications include Agri-robotics and Extreme Environments. We have developed many robotics platforms for various applications e.g. real-world application in extreme environments or research purpose platforms for the implementation of swarm scenarios inspired by social insects e.g. honeybees. Our main goal is to learn from nature, develop optimal robotic solutions, and use them to solve our problems.

Our Team

Dr Farshad Arvin

Dr Junyan Hu

Seongin Na

Shiyi Wang

Kefan Wu

Ahmad Alsayed

Songtao Xie

Zheyu Liu

Mehmet B. Bezcioglu

Yifeng He

Robotic Platforms - Equipment

Colias and Mona robots have been developed by SwaCIL in collaboration with our partners. Mona and Colias were commercialised by two SMEs in the UK.

Mona Robot

Colias

Colias

Colias-Phi

Colias-V

Mona Communication Module

TurtleBot3 Waffle Pi

Artificial Pheremone System

GPU

Research Topics

Bio-inspired Swarm Robotics

In this research theme, we investigate bio-inspired mechanisms e.g. aggregation, pheromone communication, flocking, formation, foraging etc. We model and investigate various parameters e.g. environment, individuals' behaviour, heterogeneity etc on performance of the swarm. Experiments are mostly implemented with our real-robot platforms, Colias and Mona.

Control Mechanisms from a Theoretical Perspective

In this research theme, we focus on the theoretical concept of the multi-agent system mainly focused on decentralised formation control and formation-containment scenario. We mainly model the decentralised swarm systems and investigate the effects of parameters on the swarm.

Machine Learning

In this research theme, we mainly focus on the application of machine learning in swarm robotics. We are working on the combination of bio-inspired swarm behaviours (e.g. pheromone system) and reinforcement learning to improve the performance of the swarm as well as investigating real-world applications for the swarm mechanisms.

Applications

In this research theme, we focus on the application of swarm robotics in real-world scenarios e.g. Agri-robotics and Extreme Environments. There are several ongoing projects collaborating with REEG (Robotics for Extreme Environments group) at the University of Manchester.

Mechatronics

In this research theme, we are developing hardware and robotics solutions in collaboration with our industry partners. We have close collaboration with industry in the development of micro-robots for swarm robotics and bespoke robots for extreme environments.