Swarm & Computational Intelligence Lab (SwaCIL)

SwaCIL was formed in 2019 with the main focus on Swarm Robotics and its applications in real-world scenarios. We work on bio-inspired swarm scenarios (aggregation, flocking, foraging etc) and model them with our simulation software and real micro-robots. The target applications are in Agriculture, Extreme Environment, and Cleaning.

• Control systems
• Autonomous systems 
• Mechatronics

Shiyi Wang completed his dual undergraduate degrees in mechanical electronics at Tongji University and Ernst-Abbe-Hochschule Jena in 2017. After that, he received his master's degree in electrical automation at the Technical University of Berlin in 2019. He has several years of work experience, mainly working in industrial automation, robotics, building automation, etc.

Currently, He is a Ph.D. student at the Swarm & Computational Intelligence Laboratory (SwaCIL) at the University of Manchester, UK, under the supervision of Professor Barry Lennox and Dr. Farshad Arvin. His research interests include swarm robotics and machine learning. His current research is about bio-inspired aggregation with a robot swarm using a self-built open-source platform BeeGround.

Tel. +4917684446603

Bee-Ground: an open-source simulation tool for aggregation of swarm robots controlled by the bio-inspired algorithm BEECLUST

Bee-Ground is an open-source simulation tool based on Unity and Unity Machine Learning Agents which can be applied to the research on the aggregation of swarm robots, especially the swarm robots controlled by the bio-inspired algorithm BEECLUST. MONA is the modeled robot in this simulation software, however different robotic platforms can be easily developed in Bee-Ground.

• Bee-Ground is an open-source, cross-platform simulation tool
• Bee-Ground can simulate the operation of swarm robots in various complex and dynamic environments, including obstacles and multiple heat source scenarios.
• Bee-Ground performs multi-layer multi-scenario simulations simultaneously, and the simulation speed can reach 100 times as the real-time without losing sampling resolution.
• Bee-Ground provides extended possibilities for the application of machine learning technics in swarm robotics.

The results from Bee-Ground simulation tool validate the obvious results of many previous studies that have used real robots for experiments. It has greatly improved the efficiency of swarm robotics research and has also obtained many new conclusions. Bee-Ground is an open-source tool for education. All the sources are available on GitHub.

GitHub Link: https://github.com/wshiyi-cn/BeeGround