Efficient multi-strategic hierarchical motion planning with visual servoing constraints

Most current robotic systems treat motion planning and control as two independent subsystems where the motion planner determines a collision-free path for achieving a goal, and the control system attempts to follow the planned path using sensing. This may yield undesirable motion plans that do not utilize the sensing effectively or consider sensor constraints. In this paper we consider an approach for motion planning that incorporates visual servoing constraints into the computation of the motion plans. The approach extends the notion of configuration space to include the corresponding sensor values. We address the issue of representing the extended planning space and practical computational schemes for motion planning with visual servoing constraints. We propose a hierarchical representation of the high dimensional planning space involved, and a multi-strategic heuristics search. This results in a practical motion planning scheme that is proven to be resolution-complete