DeepAI AI Chat
Log In Sign Up

Learning Deep Policies for Physics-Based Manipulation in Clutter

by   Wissam Bejjani, et al.
University of Leeds

Uncertainty in modeling real world physics makes transferring traditional open-loop motion planning techniques from simulation to the real world particularly challenging. Available closed-loop policy learning approaches, for physics-based manipulation tasks, typically either focus on single object manipulation, or rely on imitation learning, which inherently constrains task generalization and performance to the available demonstrations. In this work, we propose an approach to learn a policy for physics-based manipulation in clutter, which enables the robot to react to the uncertain dynamics of the real world. We start with presenting an imitation learning technique which compiles demonstrations from a sampling-based planner into an action-value function encoded as a deep neural network. We then use the learned action-value function to guide a look-ahead planner, giving us a control policy. Lastly, we propose to refine the deep action-value function through reinforcement learning, taking advantage of the look-ahead planner. We evaluate our approach in a physics-enabled simulation environment with artificially injected uncertainty, as well as in a real world task of manipulation in clutter.


page 1

page 6

page 7


Distilling Motion Planner Augmented Policies into Visual Control Policies for Robot Manipulation

Learning complex manipulation tasks in realistic, obstructed environment...

Asking for Help: Failure Prediction in Behavioral Cloning through Value Approximation

Recent progress in end-to-end Imitation Learning approaches has shown pr...

Learning Physics-Based Manipulation in Clutter: Combining Image-Based Generalization and Look-Ahead Planning

Physics-based manipulation in clutter involves complex interaction betwe...

Sample-Efficient Learning of Nonprehensile Manipulation Policies via Physics-Based Informed State Distributions

This paper proposes a sample-efficient yet simple approach to learning c...

Object Reconfiguration with Simulation-Derived Feasible Actions

3D object reconfiguration encompasses common robot manipulation tasks in...

Generating Stable and Collision-Free Policies through Lyapunov Function Learning

The need for rapid and reliable robot deployment is on the rise. Imitati...

Learning Closed-loop Dough Manipulation Using a Differentiable Reset Module

Deformable object manipulation has many applications such as cooking and...