Fast planning of precision grasps for three-dimensional objects

In the near future, more and more robots will be used for servicing tasks, tasks in hazardous environments or space applications. Dextrous hands are a powerful and flexible tool to interact with these real world environments that are not specially tailored for robots. In order to grasp and manipulat e real world objects, grasp planning systems are required. Grasp planning f or general three-dimensional (3-D) objects is quite a complex problem requi ring a large amount of computing time. Fast algorithms are required to inte grate grasp planners in online planning systems for robots. This paper pres ents a heuristic approach towards fast planning of precision grasps for rea listic, arbitrarily shaped 3-D objects. In this approach a number of feasib le grasp candidates are generated heuristically. These grasp candidates are qualified using an efficiently computable grasp quality measure and the be st candidate is chosen. It is shown that only a relatively small number of grasp candidates has to be generated in order to obtain a good - although n ot optimal - grasp. In Sections 1 and 2 an introduction to the problem fiel d is given and the DLR Hand is presented. Section 3 describes the grasp qua lity measure used. In Sections 4 and 5 the planning strategy is presented, i.e. how to generate and test grasp candidates. The paper closes with some results (Section 6) and an outlook (Section 7).