Robot Swarms for Surface Inspection Applications, Darren Chiu, UG '23 (3958946)
Robotic swarm inspection offers a flexible, scalable, and cost-effective solution in comparison to human inspection such as the benefit of being resilient to individual failures and simplicity in design. This research implements previous work on a distributed decision making algorithm on a swarm of three centimeter wheeled robots called Rovables. The algorithm is initially simulated through Webots, an open source physics based robotic simulator, in order to formalize and investigate the performance by determining if a randomly generated black and white grid is mostly white or black. By simulating 100 instances, we concluded that the algorithm quickly and accurately classified a grid of 55% white. Then, we moved to extend the algorithm towards vibration classification, by modifying the algorithm to determine whether more than half a metal arena was vibrating. We then found that the robots were able to successfully make a collective decision on the state of the arena as mostly vibrating or not.