Corpus ID: 232110698

ILoSA: Interactive Learning of Stiffness and Attractors

@article{Franzese2021ILoSAIL,
  title={ILoSA: Interactive Learning of Stiffness and Attractors},
  author={Giovanni Franzese and Anna M'esz'aros and Luka Peternel and Jens Kober},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.03099}
}
Teaching robots how to apply forces according to our preferences is still an open challenge that has to be tackled from multiple engineering perspectives. This paper studies how to learn variable impedance policies where both the Cartesian stiffness and the attractor can be learned from human demonstrations and corrections with a user-friendly interface. The presented framework, named ILoSA, uses Gaussian Processes for policy learning, identifying regions of uncertainty and allowing interactive… Expand
1 Citations

Figures and Tables from this paper

Teaching Robots to Grasp Like Humans: An Interactive Approach
TLDR
This work investigates how the intricate task of grasping may be learned from humans based on demonstrations and corrections, and presents a framework that learns the desired movement dynamics based on the current Cartesian position with Gaussian Processes, resulting in a reactive, time-invariant policy. Expand

References

SHOWING 1-10 OF 37 REFERENCES
Learning Collaborative Impedance-Based Robot Behaviors
TLDR
The results show that the model can be used to modify the robot impedance along task execution to facilitate the collaboration, by triggering stiff and compliant behaviors in an on-line manner to adapt to the user's actions. Expand
Force-Based Learning of Variable Impedance Skills for Robotic Manipulation
TLDR
A framework to teach a robot to perform manipulation tasks by integrating force sensing and variable impedance control is proposed, which endows robots with force-based variable stiffness skills that become relevant when vision information is unavailable or uninformative. Expand
Learning force control policies for compliant manipulation
TLDR
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. Expand
Learning Compliant Manipulation through Kinesthetic and Tactile Human-Robot Interaction
Robot Learning from Demonstration (RLfD) has been identified as a key element for making robots useful in daily lives. A wide range of techniques has been proposed for deriving a task model from aExpand
Learning variable impedance control
TLDR
The results show that the power of variable impedance control is made available to a wide variety of robotic systems and practical applications and can be used not only for planning but also to derive variable gain feedback controllers in realistic scenarios. Expand
Learning-based control strategy for safe human-robot interaction exploiting task and robot redundancies
TLDR
A control strategy for a robotic manipulator operating in an unstructured environment while interacting with a human operator is proposed, and combined with a safety strategy for tasks requiring humans to move in the vicinity of robots. Expand
Reinforcement learning for improving imitated in-contact skills
  • Murtaza Hazara, V. Kyrki
  • Computer Science
  • 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids)
  • 2016
TLDR
This work presents a framework for imitating an in-contact skill from a human demonstration and automatically enhancing the imitated force profile using a policy search method, and demonstrates the effectiveness of this approach in improving the performance of a wood planing task. Expand
Compliant skills acquisition and multi-optima policy search with EM-based reinforcement learning
TLDR
Two applications exploiting EM-based exploration strategies are presented, by considering parameterized policies based on dynamical systems, and by using Gaussian mixture models for the search of multiple policy alternatives. Expand
Geometry-aware manipulability learning, tracking, and transfer
TLDR
This article presents a novel manipulability transfer framework, a method that allows robots to learn and reproduce manipulable ellipsoids from expert demonstrations based on a tensor-based formulation of a Gaussian mixture model that takes into account that manipULability ellipSOids lie on the manifold of symmetric positive-definite matrices. Expand
Impedance Learning for Robotic Contact Tasks Using Natural Actor-Critic Algorithm
TLDR
This paper considers a learning strategy of motor skill for robotic contact tasks based on a human motor control theory and machine learning schemes that optimizes the performance of the contact tasks in uncertain conditions of the environment. Expand
...
1
2
3
4
...