Header logo is am

Learning {Motion} {Primitive} {Goals} for {Robust} {Manipulation}


Conference Paper



Applying model-free reinforcement learning to manipulation remains challenging for several reasons. First, manipulation involves physical contact, which causes discontinuous cost functions. Second, in manipulation, the end-point of the movement must be chosen carefully, as it represents a grasp which must be adapted to the pose and shape of the object. Finally, there is uncertainty in the object pose, and even the most carefully planned movement may fail if the object is not at the expected position. To address these challenges we 1) present a simplified, computationally more efficient version of our model-free reinforcement learning algorithm PI2; 2) extend PI2 so that it simultaneously learns shape parameters and goal parameters of motion primitives; 3) use shape and goal learning to acquire motion primitives that are robust to object pose uncertainty. We evaluate these contributions on a manipulation platform consisting of a 7-DOF arm with a 4-DOF hand.

Author(s): Stulp, F. and Theodorou, E. and Kalakrishnan, M. and Pastor, P. and Righetti, L. and Schaal, S.
Book Title: IEEE/RSJ International Conference on Intelligent Robots and Systems
Pages: 325--331
Year: 2011
Month: sep
Publisher: IEEE

Department(s): Autonomous Motion, Movement Generation and Control
Bibtex Type: Conference Paper (inproceedings)

DOI: 10.1109/IROS.2011.6094877

Address: San Francisco, USA
URL: https://ieeexplore.ieee.org/abstract/document/6094877/


  title = {Learning {Motion} {Primitive} {Goals} for {Robust} {Manipulation}},
  author = {Stulp, F. and Theodorou, E. and Kalakrishnan, M. and Pastor, P. and Righetti, L. and Schaal, S.},
  booktitle = {{IEEE}/{RSJ} {International} {Conference} on {Intelligent} {Robots} and {Systems}},
  pages = {325--331},
  publisher = {IEEE},
  address = {San Francisco, USA},
  month = sep,
  year = {2011},
  url = {https://ieeexplore.ieee.org/abstract/document/6094877/},
  month_numeric = {9}