Coordination and Cooperation in Multivehicle Systems
National Science Foundation CAREER Award, Air Force Office of Scientific Research FA9550-05-1-0430, The Boeing Company
In the area of coordinated control, we are interested in particular features that are most relevant to multi-vehicle systems with communication over shared media or in other ways extremely limited. Further, we are interested in vehicle’s whose dynamics incorporate realistic motion constraints such as not being able to move directly sideways or not being allowed to move in reverse. To incorporate the use of transmitted data in a restrictive media, we desire our techniques to allow for delays in communication as well as dynamic connectivity between the vehicles (in the event of lost or unavailable transmissions). In order to achieve coordinated control in this setting, we are incorporating and extending ideas from coordinated control based on Kuramoto oscillator models and ideas from stability analysis using dynamic communication graphs for linear systems. By modeling planar vehicle motion using Frenet-Serret formulas and 3D motion using natural Frenet frames, nonholonomic constraints and actuation bounds (forward velocity and turning rate) can be incorporated into vehicle dynamics. With fixed forward velocity, turning rate controls can be decomposed into a superposition of spacing and heading. Coupling the vehicle heading angles as in Kuramoto oscillator systems allows for analytical study of nonlinear coupled systems. Discretization of these models allows for the incorporation of dynamic communication, delay and asynchronicity. Shown here is an application of these ideas to cause the centroid of a group of 3D vehicles to track a dynamic moving target.
