Reinforcement Learning of Active Vision for Manipulating Objects under Occlusions

Abstract

We consider artificial agents that learn to jointly control their gripperand camera in order to reinforcement learn manipulation policies in the presenceof occlusions from distractor objects. Distractors often occlude the object of in-terest and cause it to disappear from the field of view. We propose hand/eye con-trollers that learn to move the camera to keep the object within the field of viewand visible, in coordination to manipulating it to achieve the desired goal, e.g.,pushing it to a target location. We incorporate structural biases of object-centricattention within our actor-critic architectures, which our experiments suggest tobe a key for good performance. Our results further highlight the importance ofcurriculum with regards to environment difficulty. The resulting active vision /manipulation policies outperform static camera setups for a variety of clutteredenvironments.

The paper was present in Conference of Robot Learning 2018

References in corpus (7)

• Continuous control with deep reinforcement learning
• Hindsight Experience Replay
• Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World
• Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics
• On Learning Where To Look
• Learning 6-DoF Grasping and Pick-Place Using Attention Focus
• Neural Task Programming: Learning to Generalize Across Hierarchical Tasks

Cited by in corpus (8)

• Active Perception using Light Curtains for Autonomous Driving
• Active Feature Acquisition with Generative Surrogate Models
• ROLL: Visual Self-Supervised Reinforcement Learning with Object Reasoning
• Object Finding in Cluttered Scenes Using Interactive Perception
• Active Safety Envelopes using Light Curtains with Probabilistic Guarantees
• Active 6D Multi-Object Pose Estimation in Cluttered Scenarios with Deep Reinforcement Learning
• Deep Reinforcement Learning for Active Human Pose Estimation
• An Empirical Method to Quantify the Peripheral Performance Degradation in Deep Networks