Efficient Robot Skill Learning with Imitation from a Single Video for Contact-Rich Fabric Manipulation

by   Shengzeng Huo, et al.

Classical policy search algorithms for robotics typically require performing extensive explorations, which are time-consuming and expensive to implement with real physical platforms. To facilitate the efficient learning of robot manipulation skills, in this work, we propose a new approach comprised of three modules: (1) learning of general prior knowledge with random explorations in simulation, including state representations, dynamic models, and the constrained action space of the task; (2) extraction of a state alignment-based reward function from a single demonstration video; (3) real-time optimization of the imitation policy under systematic safety constraints with sampling-based model predictive control. This solution results in an efficient one-shot imitation-from-video strategy that simplifies the learning and execution of robot skills in real applications. Specifically, we learn priors in a scene of a task family and then deploy the policy in a novel scene immediately following a single demonstration, preventing time-consuming and risky explorations in the environment. As we do not make a strong assumption of dynamic consistency between the scenes, learning priors can be conducted in simulation to avoid collecting data in real-world circumstances. We evaluate the effectiveness of our approach in the context of contact-rich fabric manipulation, which is a common scenario in industrial and domestic tasks. Detailed numerical simulations and real-world hardware experiments reveal that our method can achieve rapid skill acquisition for challenging manipulation tasks.


page 1

page 3

page 4

page 5

page 7

page 8

page 9

page 12


Learning to Manipulate Tools by Aligning Simulation to Video Demonstration

A seamless integration of robots into human environments requires robots...

Learning to Scaffold the Development of Robotic Manipulation Skills

Learning contact-rich, robotic manipulation skills is a challenging prob...

Learning by Watching: Physical Imitation of Manipulation Skills from Human Videos

We present an approach for physical imitation from human videos for robo...

Learning Skill-based Industrial Robot Tasks with User Priors

Robot skills systems are meant to reduce robot setup time for new manufa...

Learning Multimodal Contact-Rich Skills from Demonstrations Without Reward Engineering

Everyday contact-rich tasks, such as peeling, cleaning, and writing, dem...

A System for Imitation Learning of Contact-Rich Bimanual Manipulation Policies

In this paper, we discuss a framework for teaching bimanual manipulation...

Skill-based Multi-objective Reinforcement Learning of Industrial Robot Tasks with Planning and Knowledge Integration

In modern industrial settings with small batch sizes it should be easy t...

Please sign up or login with your details

Forgot password? Click here to reset