A Stochastic Approach to Generate Emergent Behaviors in Robotic Swarms

• Paulo Rezeck UFMG
• Luiz Chaimowicz UFMG

DOI: https://doi.org/10.5753/sbrlars_estendido.2024.244499

Resumo

This research introduces a novel stochastic methodology leveraging Gibbs Random Fields (GRFs) to promote a variety of emergent behaviors in robotic swarms, such as flocking, segregation, cooperative object transport, and pattern formation. By utilizing only local information and decentralized control mechanisms, the approach ensures robust and scalable swarm behaviors. Through both numerical simulations and real-world experiments using HeRo 2.0, a low-cost swarm robotic platform developed in this work, we demonstrate the benefits of the proposed methodology, including adaptability, robustness, and resilience. The stochastic approach proposed here is positioned to significantly advance the field of swarm robotics by enabling diverse behaviors, contributing to the evolution of the field, and shaping the way for new applications.

Palavras-chave: Robotics, Multi-agent system, Swarm robotics, Probabilistic robotics

Referências

Baras, J. S. and Tan, X. (2004). Control of autonomous swarms using gibbs sampling. In 2004 43rd IEEE Conference on Decision and Control (CDC)(IEEE Cat. No. 04CH37601), volume 5, pages 4752–4757. IEEE.

Fernando, M. (2021). Online flocking control of uavs with mean-field approximation. In 2021 IEEE International Conference on Robotics and Automation (ICRA), pages 8977–8983. IEEE.

Rezeck, P., Assunção, R. M., and Chaimowicz, L. (2021a). Cooperative object transportation using gibbs random fields. In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 9131–9138. IEEE.

Rezeck, P., Assunção, R. M., and Chaimowicz, L. (2021b). Flocking-segregative swarming behaviors using gibbs random fields. In 2021 IEEE International Conference on Robotics and Automation (ICRA), pages 8757–8763. IEEE.

Rezeck, P., Azpurua, H., and Chaimowicz, L. (2017). Hero: An open platform for robotics research and education. In Latin American Robotics Symposium (LARS/SBR), pages 1–6. IEEE.

Rezeck, P., Azpúrua, H., Corrêa, M. F. S., and Chaimowicz, L. (2023). Hero 2.0: a low-cost robot for swarm robotics research. Autonomous Robots.

Rezeck, P. and Chaimowicz, L. (2022). Chemistry-inspired pattern formation with robotic swarms. IEEE Robotics and Automation Letters, 7(4):9137–9144.

Schranz, M., Umlauft, M., Sende, M., and Elmenreich, W. (2020). Swarm robotic behaviors and current applications. Frontiers in Robotics and AI, 7:36.

Tan, X., Xi, W., and Baras, J. S. (2010). Decentralized coordination of autonomous swarms using parallel gibbs sampling. Automatica, 46(12):2068–2076.

Xi, W., Tan, X., and Baras, J. S. (2006). Gibbs sampler-based coordination of autonomous swarms. Automatica, 42(7):1107–1119.