Multi-robot Cooperative Control
We focus on some frontier subjects in multi-agent system to overcome the challenges of achieving advanced intelligence and security performance, including intelligent sensing, decision-making, cooperative control, and attack and defense theories, as well as their applications in industrial fields, such as environmental monitoring, target search and tracking.
HCA-based Targets Search
We introduce multi-peak PDM to model the possible positions of targets. This formulation provides an accurate model basis to the efficient targets search which improve the accuracy and efficiency of following steps. The concept of HCA is proposed to extract important regions from multi-peak PDM, and a HCA-based coordinated search method is designed to reduce time cost.
Attack against Formation
We propose an intelligent attack strategy against the robots formation with obstacle-avoidance. The attacker can infer the area by trial and error and regress the mechanism by machine learning methods. Then, an intelligent attack is designed that the formation robots are fooled to move into a preset trap.
Participated Members
Selected Publication
- Y Li, J He, C Chen, X Guan, “Learning-based Intelligent Attack against Formation Control with Obstacle-avoidance“, in Proceedings of IEEE American Control Conference, 2019.
- H Wang, Y Li, W Yu, J He, X Guan, “Moving Obstacle Avoidance and Topology Recovery for Multi-agent Systems“, in Proceedings of IEEE American Control Conference, 2019.
- Y Li, H Wang, J He, X Guan, “Optimal Topology Recovery Scheme for Multi-robot Formation Control“, in Proceedings of IEEE International Symposium on Industrial Electronics, 2019.