• Corpus ID: 239024560

Learning Robotic Manipulation Skills Using an Adaptive Force-Impedance Action Space

  title={Learning Robotic Manipulation Skills Using an Adaptive Force-Impedance Action Space},
  author={Maximilian Ulmer and Elie Aljalbout and Sascha Schwarz and Sami Haddadin},
Intelligent agents must be able to think fast and slow to perform elaborate manipulation tasks. Reinforcement Learning (RL) has led to many promising results on a range of challenging decision-making tasks. However, in real-world robotics, these methods still struggle, as they require large amounts of expensive interactions and have slow feedback loops. On the other hand, fast human-like adaptive control methods can optimize complex robotic interactions, yet fail to integrate multimodal… 

Figures and Tables from this paper


A Comparison of Action Spaces for Learning Manipulation Tasks
This paper compares learning performance across three tasks, four action spaces, and using two modern reinforcement learning algorithms to lend support to the hypothesis that learning references for a task-space impedance controller significantly reduces the number of samples needed to achieve good performance across all tasks and algorithms.
Learning force control policies for compliant manipulation
This work presents an approach to acquiring manipulation skills on compliant robots through reinforcement learning, and uses the Policy Improvement with Path Integrals (PI2) algorithm to learn these force/torque profiles by optimizing a cost function that measures task success.
Learning Variable Impedance Control for Contact Sensitive Tasks
This letter investigates how the choice of action space can give robust performance in presence of contact uncertainties and proposes learning a policy giving as output impedance and desired position in joint space and compares the performance of that approach to torque and position control under different contact uncertainties.
A Framework for Robot Manipulation: Skill Formalism, Meta Learning and Adaptive Control
A novel framework for expressing and learning force-sensitive robot manipulation skills is introduced based on a formalism that extends previous work on adaptive impedance control with meta parameter learning and compatible skill specifications and was able to perform the learned tasks without visual feedback faster than humans.
Dot-to-Dot: Explainable Hierarchical Reinforcement Learning for Robotic Manipulation
A hierarchical deep reinforcement learning system, consisting of a low-level agent handling the large actions/states space of a robotic system efficiently, by following the directives of a high- level agent which is learning the high-level dynamics of the environment and task.
Hierarchical Reinforcement Learning for Robot Navigation
A HRL architecture for learn- ing robot's movements, e.g. robot navigation, which consists of two layers: movement planning and movement execution and which is implemented and evaluated on a mobile robot platform for a navigation task.
Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks
It is shown that VICES improves sample efficiency, maintains low energy consumption, and ensures safety across all three experimental setups, and RL policies learned with VICES can transfer across different robot models in simulation, and from simulation to real for the same robot.
Policy search for motor primitives in robotics
A novel EM-inspired algorithm for policy learning that is particularly well-suited for dynamical system motor primitives is introduced and applied in the context of motor learning and can learn a complex Ball-in-a-Cup task on a real Barrett WAM™ robot arm.
Layered direct policy search for learning hierarchical skills
A new HRL algorithm based on information theoretic principles to autonomously uncover a diverse set of sub-policies and their activation policies is proposed and evaluated on two high dimensional continuous tasks.
Hierarchical Reinforcement Learning for Quadruped Locomotion
The first application of end-to-end hierarchical learning to a real robotic locomotion task and it is shown that steering behaviors automatically emerge in the latent command space as low-level skills are needed for this task.