Introduction

The interest in the research of cooperative control of Multi Agent Systems (MAS) has been increased in the last decades. This is stimulated by the technological advances and the development of relatively inexpensive communication, computation, and sensing devices. In multi-agent systems, each individual is assumed to have abilities to sense its immediate environment, communicate with others, process the information gathered and take a local action based on the information gathered.

Key aspects of designing the control laws for motion coordination are that each agent has to behave without any leader (leader-less) and based on its local information (i.e. the control law is distributed) due to the limited communication and sensing capabilities. This is also inspired by the bahavior of biological systems such as school of fish, flocking of birds and animal herds that exist in nature.

Research statement

We are interested in cooperative control of a team consists of humans and robots in joint manipulation or performing other tasks. The goal is to have the robots cooperate synchronously and smoothly with the human when they have to execute a task together. This brings new challenges in the area of cooperative control of MAS due to the complex dynamics of the human. This research will combine results from control theory, graph theory, computer science together with psychological studies of the human model.

Some challenges addressed in this research are:

• Stability and performance analysis in the presence of agents with stochastic dynamic system models
• Network topology (information flow)