Distributed, vision -based control laws for motion coordination in multi-agent systems

Nima Moshtagh, University of Pennsylvania


From ecology and evolutionary biology to social sciences, and from systems and control theory to statistical physics and computer graphics, researchers have been trying to develop an understanding of how a group of moving objects such as flocks of birds, schools of fish and crowds of people can perform collective tasks such as reaching a consensus or moving in a formation without centralized coordination. Researchers in the fields of robotics and control theory have also become interested in cooperative control of multiple autonomous robots because of tremendous military and civilian applications. ^ First, novel ways are introduced to generate collective behaviors within a team of mobile robots using only visual sensing. The proposed distributed control laws do not rely on the communication or measurement of heading or distance information among neighbors, but instead require measurements of bearing, optical flow and time-to-collision, all of which can be measured using simple vision sensors. The effectiveness of the control laws are demonstrated on a group of mobile robots.^ Secondly, the consensus approach is used to generate distributed motion coordination control laws for a team of robots with dynamics of each robot evolving on SE(3) (flying formations) and SO(3) (satellite attitudes). A series of distributed control laws for collective behaviors such as parallel, circular and line formations of a team of robots are presented. Using Lyapunov techniques the stability of relative equilibria are studied. In all cases, ideas from graph theory and control theory are merged to synthesize the desired control laws and analyze their stability.^

Subject Area

Engineering, Robotics

Recommended Citation

Moshtagh, Nima, "Distributed, vision -based control laws for motion coordination in multi-agent systems" (2008). Dissertations available from ProQuest. AAI3328697.