SWARM's Opportunity Design

Advancements in swarm robotics can revolutionize fields such as disaster recovery, agriculture, and autonomous exploration. Perhaps the biggest challenge to implementing swarms is understanding how humans and robots can collaborate effectively in dynamic environments. The Simulator for Human and Swarm Team Applications (SHaSTA) was designed as an experimental system to study human-swarm interaction. The initial version is not realistic enough for intricate investigations. This capstone aims to enhance SHaSTA with Virtual Reality (VR), Robot Operating System (ROS 2), a more user-friendly interface, and eventually include a Lab Streaming Layer (LSL) to capture physiological signals. These enhancements will render the simulation environment more immersive, flexible, and interactive.

This capstone project seeks to enhance SHaSTA's capabilities through the integration of state-of-the-art technologies that increase realism and responsiveness into the simulator. By adding Virtual Reality (VR), ROS 2, Gazebo, and Lab Streaming Layer, SHaSTA will be a more effective research tool for studying human-swarm interaction under realistic conditions. These enhancements will enable enhanced situational awareness, real-time swarm control, and adaptive behavior tuning, giving users a more functional and intuitive interface.

The Virtual Reality (VR) component will provide an immersive perspective, allowing users to interact with swarm robots in a 3D environment. This feature enhances situational awareness and enables more intuitive interaction with the swarm. By visualizing robots and their surroundings in a spatially realistic manner, operators can make more informed decisions, improving coordination and efficiency in multi-agent tasks.

ROS 2 integration will be the backbone of communication in the system, facilitating smooth real-time exchange of data between human operators and swarm robots. ROS 2 offers modular, scalable, and communication that enables swarm members to share sensor information, issue commands, and dynamically negotiate behavior. This kind of infrastructure is vital in enabling a responsive simulation environment where there is minimal latency involved in interactions between robots and humans.

The Gazebo simulation platform will be used as the primary interface for testing and controlling of the swarm. Gazebo is a physics-based virtual testbed that allows users to watchand control robotic swarms with an intuitive web-based interface. Integration allows operators to execute complicated missions, change parameters, and monitor robot behavior in real-time. Gazebo's ability to run Python scripts and ROS-based commands will make it a powerful tool for simulation and testing, reducing the amount of additional user interface coding.

Finally, physiological data collection using Lab Streaming Layer (LSL) will introduce an adaptive element to swarm behavior. LSL shall be utilized for capturing real-time biometric measurements such as heart rate and EEG signals so the system can alter swarm behavior in relation to the operator's cognitive load and level of stress. By this closed-loop feedback process, the swarm can dynamically compensate for the state of mind of the user and reduce task difficulty when stressed and maximize automation whenever cognitive load crosses the threshold. While this ability can be achieved in later stages of development, it is a significant step toward human-aware and responsive swarms.

These enhancements will significantly increase SHaSTA's realism and usability, making it an even more powerful research tool. By seeing how physiological states influence human swarm cooperation, researchers can develop more adaptive and efficient swarm control strategies, with robots operating in perfect coordination with human operators in a way that is intuitive, responsive, and effective.