Solving precision grasp with simple feature-matching technique

We present an improved strategy for finding kinematically-feasible precision grasps for particular robot hand. Following the concept of collecting samples of hand posture and classifying them into a database of information feature representing each hand configuration in our previous work, we can efficiently search for valid hand postures given a set of contact points with improvement in much faster processing time compared to the last one. This newly-designed approach for creating hand posture database can still be applied to any kind of robot hands as long as the kinematic specification of those hands are provided.