Exportar Publicação

Romano, P., Nunes, L. & Oliveira, S. (2023).  Hybrid training to generate robust behaviour for swarm robotics tasks. In van Stein, N., Marcelloni, F., Lam, H. K., Cottrell, M., and Filipe, J. (Ed.), Proceedings of the 15th International Joint Conference on Computational Intelligence. (pp. 265-277). Rome, Italy: SciTePress.

P. S. Romano et al.,  " Hybrid training to generate robust behaviour for swarm robotics tasks", in Proc. of the 15th Int. Joint Conf. on Computational Intelligence, van Stein, N., Marcelloni, F., Lam, H. K., Cottrell, M., and Filipe, J., Ed., Rome, Italy, SciTePress, 2023, pp. 265-277

@inproceedings{romano2023_1732258422642,
	author = "Romano, P. and Nunes, L. and Oliveira, S.",
	title = " Hybrid training to generate robust behaviour for swarm robotics tasks",
	booktitle = "Proceedings of the 15th International Joint Conference on Computational Intelligence",
	year = "2023",
	editor = "van Stein, N., Marcelloni, F., Lam, H. K., Cottrell, M., and Filipe, J.",
	volume = "",
	number = "",
	series = "",
	doi = "10.5220/0012193300003595",
	pages = "265-277",
	publisher = "SciTePress",
	address = "Rome, Italy",
	organization = "",
	url = "https://www.scitepress.org/ProceedingsDetails.aspx?ID=z5f/p31NVmI=&t=1"
}

TY  - CPAPER
TI  -  Hybrid training to generate robust behaviour for swarm robotics tasks
T2  - Proceedings of the 15th International Joint Conference on Computational Intelligence
AU  - Romano, P.
AU  - Nunes, L.
AU  - Oliveira, S.
PY  - 2023
SP  - 265-277
DO  - 10.5220/0012193300003595
CY  - Rome, Italy
UR  - https://www.scitepress.org/ProceedingsDetails.aspx?ID=z5f/p31NVmI=&t=1
AB  -  Training of robotic swarms is usually done for a specific task and environment. The more specific the training is, the more the likelihood of reaching a good performance. Still, flexibility and robustness are essential for autonomy, enabling the robots to adapt to different environments. In this work, we study and compare approaches to robust training of a small simulated swarm on a task of cooperative identification of moving objects. Controllers are obtained via evolutionary methods. The main contribution is the test of the effectiveness of training in multiple environments: simplified versions of terrain, marine and aerial environments, as well as on ideal, noisy and hybrid (mixed environment) scenarios. Results show that controllers can be generated for each of these scenarios, but, contrary to expectations, hybrid evolution and noisy training do not, in general, generate better controllers for the different scenarios. Nevertheless, the hybrid controller reaches a performance level par with specialized controllers in several scenarios, and can be considered a more robust solution.
ER  -