Abstract-Motion coordination of autonomous vehicles has applications from target surveillance to climate monitoring. Previous research has yielded stabilizing control laws for a selfpropelled-vehicle model with first-order rotational dynamics; however, this model may not adequately describe the rotational dynamics of vehicles in the atmosphere or ocean. This paper describes the design of backstepping algorithms for the decentralized control of self-propelled vehicles with secondorder rotational dynamics. We design backstepping controls for planar parallel and circular formations in the absence of a flowfield and in the presence of a steady, uniform flowfield. These controls extend prior results to a more realistic vehicle model.

Abstract-Collective motion of a multi-vehicle testbed has applications in weather monitoring and ocean sampling. Previous work in this field has produced theoretically justified algorithms for stabilization of parallel and circular motions of self-propelled particles using measurements of relative position and relative velocity. This paper describes an observer-based feedback algorithm for stabilization of parallel and circular motions using measurements of relative position only. This algorithm utilizes information about the particle dynamics and turning rates to estimate the relative velocities. We describe a laboratory-scale underwater vehicle testbed on which the algorithm is being implemented.

Multi-vehicle control has applications in weather monitoring and ocean sampling. Previous work in this field has produced theoretically justified algorithms for stabilization of parallel and circular motions of self-propelled Newtonian particles using measurements of relative position and relative velocity. This paper describes an observer-based feedback control algorithm for stabilization of parallel and circular motions using measurements of relative position only. The algorithm utilizes information about vehicle dynamics and turning rates to estimate relative velocity.Theoretical justification is provided for the vehicle model, and numerical simulations suggest that the algorithm extends to a three-dimensional rigid body model. The algorithm has been implemented on a laboratory-scale underwater vehicle testbed, and we describe the results of experimental validation in the University of Maryland's Neutral Buoyancy Research Facility.

The Synthetic Collective Unmanned Underwater Laboratory (SCUUL) testbed is a multi-vehicle testbed that is used to evaluate the performance of underwater motion coordination algorithms in a dynamic environment. The SCUUL testbed consists of six propellor-driven vehicles, a 367,000 gallon tank, and an underwater motion capture system. The tank is the Neutral Buoyancy Research Facility (NBRF) located at the University of Maryland and operated by the Collective Dynamics and Control Laboratory (CDCL) and the Space Systems Laboratory. The motion capture is a state-of-the-art system developed by Qualisys in Gothenburg, Sweden. Initial results have shown the capabilities of the separate components of SCUUL and its ability to test and evaluate a multitude of motion coordination algorithms in a laboratory environment.