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 :
Preprints, Working Papers, ...
Complete list of metadatas

Cited literature [28 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01820043
Contributor : François de la Bourdonnaye <>
Submitted on : Thursday, June 21, 2018 - 11:16:09 AM
Last modification on : Saturday, July 7, 2018 - 1:08:36 AM
Long-term archiving on : Monday, September 24, 2018 - 7:30:03 PM

File

fdlb_paper_iros_v1.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01820043, version 1

Citation

François de la Bourdonnaye, Céline Teulière, Jochen Triesch, Thierry Chateau. Learning to touch objects through stage-wise deep reinforcement learning. 2018. ⟨hal-01820043⟩

Share

Metrics

Record views

582

Files downloads

607