Learning to touch objects through stage-wise deep reinforcement learning

Abstract : Learning complex behaviors through reinforcement learning is particularly challenging when reward is only available upon successful completion of the full behavior. In manipulation robotics, so-called shaping rewards are often used to overcome this problem. However, these usually require human engineering or (partial) world models describing, e.g., the kinematics of the robot or high-level modules for perception. Here we propose an alternative method to learn an object palm-touching task through a weakly-supervised and stage-wise learning of simpler tasks. First, the robot learns to fixate the object with its cameras. Second, the robot learns eye-hand coordination by learning to fixate its end effector. Third, using the previously acquired skills an informative shaping reward can be computed which facilitates efficient learning of the object palm-touching task. We demonstrate in simulation that learning the full task with this shaping reward is comparable to learning with an informative supervised reward.
Document type :
Conference papers
Complete list of metadatas

https://hal-clermont-univ.archives-ouvertes.fr/hal-01915423
Contributor : Pascale Dugat <>
Submitted on : Tuesday, January 15, 2019 - 6:18:10 PM
Last modification on : Thursday, February 7, 2019 - 3:34:47 PM

File

fdlb_paper_iros_v1.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01915423, version 1

Citation

François de La Bourdonnaye, Céline Teulière, Jochen Triesch, Thierry Chateau. Learning to touch objects through stage-wise deep reinforcement learning. IEEE/RSJ International Conference on Intelligent Robots and Systems, Oct 2018, Madrid, Spain. ⟨hal-01915423⟩

Share

Metrics

Record views

38

Files downloads

41