Mixed reality-based simulation platform for swarm robotics

Swarm intelligence is the collective behaviour of decentralized and self-organizing autonomous agents that cooperate to accomplish complex tasks. This behaviour is inspired by natural systems such as ant colonies, bee swarms, bird flocks, and other social organisms. This project was developed to address one of the major challenges in swarm robotics research, which is the high cost and complexity associated with developing and testing swarm intelligence algorithms using a large number of physical robots.

The project aims to provide a mixed reality-based simulation platform where physical and virtual robots coexist within a unified environment. Instead of requiring researchers to build or purchase large fleets of robots, the platform enables swarm algorithms to be developed, tested, and evaluated using a combination of real robotic units and virtual agents. This significantly reduces the hardware required while preserving realistic interactions between robots.

The platform provides a centralized environment for managing swarm experiments by allowing users to position robots within the simulation arena, execute swarm intelligence algorithms, monitor robot status, and visualize robot behaviour in real time. It also supports remote experimentation, enabling researchers to upload their own algorithms, deploy them to the robotic platform, and observe execution through live data and video feedback. This approach allows researchers to validate swarm behaviours before deploying algorithms on larger physical robotic systems.

The project involved developing software modules for robot communication, simulation, visualization, and synchronization, enabling physical and virtual agents to collaborate within the same mixed reality environment. The framework was designed to support distributed swarm intelligence experiments, allowing researchers to evaluate navigation, coordination, formation control, and cooperative behaviours efficiently before real-world deployment.

The platform demonstrates how mixed reality can bridge the gap between simulation and physical deployment by providing a scalable and cost-effective environment for swarm robotics research. Through this project, I gained practical experience in mixed reality technologies, robotics simulation, distributed systems, real-time communication, and software engineering for multi-agent robotic applications.

Key Responsibilities
* Designed and developed a user-friendly graphical interface for remotely supervising multiple autonomous ground robots.
* Implemented real-time command execution to control and coordinate robot movements from a centralized control panel.
* Developed monitoring features to visualize robot status, connectivity, and operational information.
* Integrated communication modules enabling reliable interaction between the control application and distributed robotic units.
* Supported collaborative swarm-robotics experiments by providing centralized monitoring and coordination capabilities.
* Participated in system integration, testing, debugging, and validation to ensure reliable operation in real-world environments.
* Collaborated with multidisciplinary teams working on swarm intelligence, robot communication, and autonomous systems.
* Contributed to improving the usability and scalability of the control platform for research and educational applications.

Technologies
Java, MQTT, Distributed Systems, Multi-Agent Systems, Robotics, GUI Development, Network Communication, Git

Gallery

Team

Dilshani Karunarathna

Nuwan Jaliyagoda

Pasan Tennakoon

Resources/ Links

Back Next