Path planning of an AFM-based nanomanipulator using virtual force reflection

This paper proposes a practical planning system for 2D assembly tasks at the microscale. To improve the real-time observation and manipulation, we proposed real-time 3-D reconstruction of the microworld using image processing and virtualized reality techniques. Then, a planner covers a whole range of problems in object assignment, obstacle detection and avoidance and path trajectory. We describe algorithms based on optimization theory and Voronoi graph construction. As automatic motion planners fail due to the difficulty of discovering critical configurations due to microdomain interactions, cooperation with operator skills is able to solve a motion planning query. The feasible haptically-generated paths based on potentials fields reaction forces are selected for efficient pushing-based manipulation without collisions.