A Virtual velocity attractor, harmonic potential approach for joint planning and control of a UAV

The objective of this work is to develop a practical, closed-loop and simple navigation controller that suits a large variety of unmanned aerial vehicles (UAVs). The method indirectly controls the trajectory of a UAV by regulating its velocity using as a reference a dense vector field derived from the gradient of a harmonic potential filed (HPF). The field functions to inject the robot's context in the control process by forcing its group structure to observe a set of state and differential constraints that reflect the contents of the environment, the goal to be reached and the constraints on behavior. The velocity regulation process is carried-out using a novel concept called the: virtual velocity attractor (VVA). The HPF approach, the VVA procedure and a recently suggested two stage approach for modeling the motion of rigid, nonholonomic robots [43] seems to be well-matched to each other enabling easy, on-line conversion of the provably-correct guidance signal from the HPF planner into a well-behaved control signal that can be fed to the actuator of the UAV.