A testbed for multi-robot systems

Abstract

<p>This thesis investigates the problem of multiple robot path planning. In the first chapter of the thesis, we propose a general purpose multi-robots testbed Cy-Eye. Typical applications include target detection, tracking, and surveillance can be tested on this testbed. Its architecture makes it suitable for centralized and distributed experiments.</p> <p>In the second chapter, we present one formation control problem. When multiple robots working together, it is often that they have to assign targets among themselves, and then plan and schedule their collision-free paths to their targets. Specifically, we present a navigation strategy for multiple ground-based robots in row crop field. We show that obtaining the solution to the problem of minimizing the length of the distance traveled by the robots, and subsequent rearrangement can lead to paths on which the robots only collide at a few intersections. Controlling the passage of robots at those intersections with local interactions can lead to collision-free paths.</p> <p>In the third chapter, we present the current progress of a multiple player pursuit-evasion game. The objects for the aerial pursuers are maximizing the tracking time for keep multiple evaders in the field of view. We propose a tracking strategy and show the simulation.</p>