• Corpus ID: 199405578

Combining learned skills and reinforcement learning for robotic manipulations

  title={Combining learned skills and reinforcement learning for robotic manipulations},
  author={Robin Strudel and Alexander Pashevich and Igor Kalevatykh and Ivan Laptev and Josef Sivic and Cordelia Schmid},
Manipulation tasks such as preparing a meal or assembling furniture remain highly challenging for robotics and vision. The supervised approach of imitation learning can handle short tasks but suffers from compounding errors and the need of many demonstrations for longer and more complex tasks. Reinforcement learning (RL) can find solutions beyond demonstrations but requires tedious and task-specific reward engineering for multi-step problems. In this work we address the difficulties of both… 

Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control

It is shown that training vision-based control policies in simulation while gradually increasing the difficulty of the task via ACGD improves the policy transfer to the real world.

Vision-Based One-Shot Imitation Learning Supplemented with Target Recognition via Meta Learning

An end-to-end meta imitation learning method supplemented with target recognition (TaR-MIL) is proposed for one-shot learning that manages to perform placing tasks in new scenarios or with new objects after one video demonstration is given, which verify the effectiveness of the proposed method.

Machine Learning for Robotic Manipulation

This document surveys recent robotics conferences and identifies the major trends with which machine learning techniques have been applied to the challenges of robotic manipulation.

Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving

A hierarchical reinforcement and imitation learning approach that consists of low-level policies learned by IL for discrete driving modes, and a high-level policy learned by RL that switches between different driving modes that can achieve higher efficiency and safety compared to other methods.



Reinforcement and Imitation Learning for Diverse Visuomotor Skills

This work proposes a model-free deep reinforcement learning method that leverages a small amount of demonstration data to assist a reinforcement learning agent and trains end-to-end visuomotor policies that map directly from RGB camera inputs to joint velocities.

Deep Reinforcement Learning for Robotic Manipulation

It is demonstrated that a recent deep reinforcement learning algorithm based on offpolicy training of deep Q-functions can scale to complex 3D manipulation tasks and can learn deep neural network policies efficiently enough to train on real physical robots.

Asymmetric Actor Critic for Image-Based Robot Learning

This work exploits the full state observability in the simulator to train better policies which take as input only partial observations (RGBD images) and combines this method with domain randomization and shows real robot experiments for several tasks like picking, pushing, and moving a block.

Overcoming Exploration in Reinforcement Learning with Demonstrations

This work uses demonstrations to overcome the exploration problem and successfully learn to perform long-horizon, multi-step robotics tasks with continuous control such as stacking blocks with a robot arm.

Composing Complex Skills by Learning Transition Policies

This work proposes a method that can learn transition policies which effectively connect primitive skills to perform sequential tasks without handcrafted rewards, and introduces proximity predictors which induce rewards gauging proximity to suitable initial states for the next skill.

Data-efficient Deep Reinforcement Learning for Dexterous Manipulation

The results show that by making extensive use of off-policy data and replay, it is possible to find control policies that robustly grasp objects and stack them and hint that it may soon be feasible to train successful stacking policies by collecting interactions on real robots.

Stochastic Neural Networks for Hierarchical Reinforcement Learning

This work proposes a general framework that first learns useful skills in a pre-training environment, and then leverages the acquired skills for learning faster in downstream tasks, and uses Stochastic Neural Networks combined with an information-theoretic regularizer to efficiently pre-train a large span of skills.

End-to-End Training of Deep Visuomotor Policies

This paper develops a method that can be used to learn policies that map raw image observations directly to torques at the robot's motors, trained using a partially observed guided policy search method, with supervision provided by a simple trajectory-centric reinforcement learning method.

One-Shot Imitation Learning

A meta-learning framework for achieving one-shot imitation learning, where ideally, robots should be able to learn from very few demonstrations of any given task, and instantly generalize to new situations of the same task, without requiring task-specific engineering.