A master-slave fluid cooperative control algorithm for optimal trajectory planning

A control algorithm based on Smoothed Particle Hydrodynamics (SPH) is used to guide a simulated swarm of unmanned vehicles along a pre-computed optimal trajectory. By using a fluid based control method, collision avoidance is automatically incorporated into the control. Given an initial position and goal location, it is computationally impractical to compute the optimal trajectories for a large swarm of vehicles. To overcome this, the optimal trajectory is computed for a single vehicle and the swarm is guided along this trajectory using fluid based control, in this case SPH. Importantly, this method avoids computing a potential field to guide the swarm and is easily adaptable to changing environments. Additionally, this method is intended to be used in the presence of a background flow through which the vehicles must travel. This control algorithm is found to be quite robust in the test case examined here and adding additional vehicles to the swarm has only a small impact on fuel consumption per vehicle.