Fuzzy Neuro Control For The Displacement Of A Swarm Of Robots In The Robotarium Environment

Swarm robotics presents challenges in trajectory control due to the complex paths followed by individual robots. This study addresses the issue by proposing a fuzzy neuro-control system that provides improved performance for each robot. The goal is to achieve successful trajectory tracking in a collaborative robot environment; however, the system's effectiveness relies on a finite and controlled setting. An impedance-based learning approach is integrated to enhance robot interaction and detect external elements. Artificial intelligence methodologies are employed for planning, control, and trajectory tracking in a dynamic environment. The methodology includes designing specific and finite trajectories, training the neuro-fuzzy system based on the leader's trajectory, and evaluating follower robot behaviors with different trajectories and distances from the leader using Georgia Tech's Robotarium platform. The platform offers a user-friendly interface, simulation capabilities, short response times, and free access for real-time algorithm implementation. The results validate the effectiveness of the proposed approach, considering the limitations of the control used, while highlighting its potential for swarm robotics research.