A fast and robust grasp planner for arbitrary 3D objects

Abstract
In the near future, more and more robots will be used for servicing tasks, tasks in hazardous environ- ments or space applications. Dextrous hands are a powerful and flexible tool to interact with real world environments that are not specially tailored for robots. In order to grasp and manipulate real world objects, grasp planning systems are required. To be integrated in online planning systems for robots, they have to be very fast. This paper presents a method to compute a desirable grasp quality measure very fast and accu- rate - both aims haven't been reached simultaneously until now. Based on this measure an heuristic ap- proach towards fast planning of precision grasps for arbitrarily shaped 3D objects is described. A num- ber of feasible grasp candidates are generated heuris- tically. These grasp candidates are qualified using the described grasp quality measure and the best candi- date is chosen. The planned grasps are robust in re- spect of grasp placement. It is shown that only a rel- atively small number of grasp candidates has to be generated in order to obtain a good - although not optimal - grasp.

This publication has 8 references indexed in Scilit: