澳门2138cn太阳娱乐网站官网-在线登录

报告内容摘要：
Use of cooperative formation of agents enables achieving goals that would not be possible for single agent. The ability to achieve, maintain and change formation is one of the fundamental prerequisites for coordinated behavior. So far various approaches to formation control are proposed and used. They vary from stochastic to deterministic control, both, open loop and closed loop, decentralized and centralized. A variety of mathematical theories such as graph theory, game theory and dynamical systems theory are therein utilized. In this talk we present three approaches to formation control in multi-agent systems:
- fuzzy logic based control,
- artificial potential based control,
- grid-based control algorithm.
Fuzzy logic based approach represents a methodology for control and decentralized decision making in multi agent systems. Driving forces are described by the set of fuzzy logic rules that have been determined by fuzzy Lyapunov stability criteria. The second approach is based on the artificial potential function that characterizes potential of the entire formation, created by the set of the bell-shaped elementary potential functions. The special feature of the introduced potential function is its dependence on a control parameter that widens or, conversely, concentrates the bell-shape changing the effective range of interactions. Finally, grid-based control algorithm is based on abstracting the agent formation by a rectangular grid. We develop a coordination strategy that ensures collision-free transitions between arbitrarily defined formations. This coordination strategy is based on a well-established control design methodology for manufacturing systems. Group navigation is achieved by integrating a waypoint planner with a novel grid-based formation reconfiguration controller. The proposed hybrid architecture is capable of driving a group of agents through a cluttered environment, while maintaining specified formation constraints. We present simulation and experimental results verifying the validity of the proposed methodologies.

报告人简介：
Stjepan Bogdan received his Ph.D.E.E. in 1999, M.S.E.E. in 1993 and B.S.E.E. in 1990 at the University of Zagreb, Croatia. Currently he is an associate professor at the Faculty of Electrical Engineering and Computing, University of Zagreb. His main areas of interest are discrete event systems, intelligent control systems, and autonomous systems. He is a coauthor of three books and numerous papers published in journals and proceedings. He is a member of IEEE CSS Technical Committee of Intelligent Control and is appointed General Chair of 2012 IEEE Multi-conference on Systems and Control. He is Vice-President of Croatian Robotics Society. He serves as associate editor of IEEE Transactions of Automation Science and Engineering, Journal of Intelligent and Robotic Systems, Transactions of the Institute of Measurement & Control and Journal of Control Theory and Applications.