Visual guided grasping and generalization using self-valuing learning

We present a self-valuing learning technique which is capable of learning how to grasp unfamiliar objects and generalize the learned abilities. The learning system consists of two learners which distinguish between local and global grasping criteria. The local criteria are not object specific while the global criteria cover physical properties of each object. In this case we present a generalization method of the learning parameters based on a tree distance model for the medial axis transformations. The system is self-valuing, i.e. it rates its actions by evaluating sensory information and the usage of image processing techniques. An experimental setup consisting of a PUMA-260 manipulator, equipped with a hand-camera and a force/torque sensor was used to test this scheme. The system has shown the ability to grasp a wide range of objects and to apply previously learned knowledge to new objects.

[1]  Jianwei Zhang,et al.  Visual guided grasping of aggregates using self-valuing learning , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[2]  Jianwei Zhang,et al.  Visuelles Reinforcement-Lernen zur Feinpositionierung eines Roboterarms über kompakte Zustandskodierung , 1999, AMS.

[3]  Antonio Morales,et al.  Heuristic vision-based computation of planar antipodal grasps on unknown objects , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[4]  Shimon Edelman,et al.  Learning to grasp using visual information , 1996, Proceedings of IEEE International Conference on Robotics and Automation.

[5]  Richard Franklin Riesenfeld,et al.  Applications of b-spline approximation to geometric problems of computer-aided design. , 1973 .

[6]  Vijay Kumar,et al.  Robotic grasping and contact: a review , 2000, Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No.00CH37065).

[7]  W. J. Gordon,et al.  B-SPLINE CURVES AND SURFACES , 1974 .

[8]  I. J. Schoenberg.  Contributions to the problem of approximation of equidistant data by analytic functions. Part A. On the problem of smoothing or graduation. A first class of analytic approximation formulae , 1946 .

[9]  I. J. Schoenberg.  Contributions to the problem of approximation of equidistant data by analytic functions. Part B. On the problem of osculatory interpolation. A second class of analytic approximation formulae , 1946 .

[10]  Lusheng Wang,et al.  Alignment of trees: an alternative to tree edit , 1995 .

[11]  Kenneth Y. Goldberg,et al.  Computing parallel-jaw grips , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[12]  Kuo-Chung Tai,et al.  The Tree-to-Tree Correction Problem , 1979, JACM.