Learning Pneumatic Non-Prehensile Manipulation with a Mobile Blower

04/05/2022
by   Jimmy Wu, et al.
0

We investigate pneumatic non-prehensile manipulation (i.e., blowing) as a means of efficiently moving scattered objects into a target receptacle. Due to the chaotic nature of aerodynamic forces, a blowing controller must (i) continually adapt to unexpected changes from its actions, (ii) maintain fine-grained control, since the slightest misstep can result in large unintended consequences (e.g., scatter objects already in a pile), and (iii) infer long-range plans (e.g., move the robot to strategic blowing locations). We tackle these challenges in the context of deep reinforcement learning, introducing a multi-frequency version of the spatial action maps framework. This allows for efficient learning of vision-based policies that effectively combine high-level planning and low-level closed-loop control for dynamic mobile manipulation. Experiments show that our system learns efficient behaviors for the task, demonstrating in particular that blowing achieves better downstream performance than pushing, and that our policies improve performance over baselines. Moreover, we show that our system naturally encourages emergent specialization between the different subpolicies spanning low-level fine-grained control and high-level planning. On a real mobile robot equipped with a miniature air blower, we show that our simulation-trained policies transfer well to a real environment and can generalize to novel objects.

READ FULL TEXT

page 1

page 2

page 4

page 6

research
05/30/2023

AlphaBlock: Embodied Finetuning for Vision-Language Reasoning in Robot Manipulation

We propose a novel framework for learning high-level cognitive capabilit...
research
12/14/2021

Fast Footstep Planning on Uneven Terrain Using Deep Sequential Models

One of the fundamental challenges in realizing the potential of legged r...
research
03/16/2023

Efficient Learning of High Level Plans from Play

Real-world robotic manipulation tasks remain an elusive challenge, since...
research
03/23/2019

Long Range Neural Navigation Policies for the Real World

Learned Neural Network based policies have shown promising results for r...
research
03/03/2020

Hierarchically Decoupled Imitation for Morphological Transfer

Learning long-range behaviors on complex high-dimensional agents is a fu...
research
05/24/2021

Coarse-to-Fine for Sim-to-Real: Sub-Millimetre Precision Across the Workspace

When training control policies for robot manipulation via deep learning,...
research
12/08/2015

Reinforcement Control with Hierarchical Backpropagated Adaptive Critics

Present incremental learning methods are limited in the ability to achie...

Please sign up or login with your details

Forgot password? Click here to reset