The work resulted in a paper published as

Farbod Fahimi, Rineesh Siddareddy and C. Nataraj, 2005, Intelligent Ships Symposium, May.

The problem of control and coordination for many unmanned surface vehicles moving in formation is investigated. The overall motion plan for the formation can be determined, for example, by using the potential field method as a trajectory planner. This motion plan is then used to control a single unmanned lead vehicle. It is assumed that each vehicle has the ability to measure the relative position of other vehicles that are immediately adjacent to it. Once the motion for the lead vehicle is given, the reminder of the formation is governed by local (decentralized) control laws based on the relative dynamics of each of the follower vehicles and the relative positions of the vehicles in the formation. The dynamics model of the vehicles consists of surge, sway, and yaw degrees of freedom. It is assumed that two thrusters drive the unmanned vehicles. Two scenarios are described for feedback control within a formation. In the first scenario, one vehicle controls its relative distance and orientation with respect to a neighboring vehicle. This situation is applicable to all formations in which each vehicle sees one neighbor except for the lead vehicle. Thus it can be used for vehicles marching in a single file. In the second scenario, a vehicle maintains its position in the formation by maintaining specified distances from two neighboring vehicles, or from one vehicle and an obstacle. When vehicles are constrained by more than one neighbor (or obstacle) in the formation, this control law becomes useful. An example for this scenario is when a vehicle is marching in a rectangular formation with some rows and columns, or when a vehicle in a single file formation is marching very near an obstacle. The methods that are proposed for controlling the formation use only local sensor-based information. Feedback linearization method is used and it is shown that the relative distances and orientation of vehicles are exponentially stabilized. The stability of zero dynamics of the system is also discussed. These control laws have the advantage of providing easily computable, real-time feedback control, with provable performance for the entire system. Several numerical simulations are presented to demonstrate the use of these techniques including motion of a group of unmanned vehicles. Finally, some concluding remarks are made about stability and sensitivity to parameter uncertainties.