Learning object-level impedance control for robust grasping and dexterous manipulation

By Miao Li, Hang Yin, K. Tahara, Aude Gemma Billard. 2014

101 citations recorded by the source corpus.

Object-level impedance control is of great importance for object-centric tasks, such as robust grasping and dexterous manipulation. Despite the recent progress on this topic, how to specify the desired object impedance for a given task remains an open issue. In this paper, we decompose the object's impedance into two complementary components-the impedance for stable grasping and impedance for object manipulation. Then, we present a method to learn the desired object's manipulation impedance (stiffness) using data obtained from human demonstration. The approach is validated in two tasks, for robust grasping of a wine glass and for inserting a bulb, using the 16 degrees of freedom Allegro Hand mounted with the SynTouch tactile sensors.

Subfields: Biomedical Engineering, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Control and Systems Engineering

Citations in the corpus

Solving Rubik's Cube with a Robot Hand (2019, neutral): “tactile sensing [104, 59, 60, 61].”
Learning robot in-hand manipulation with tactile features (2015, methodology): “Apart from providing information on the object’s pose and contact normals [11–17], tactile sensing can provide greater robustness to variations in object properties [18–21], perturbations [22, 23], and sensing errors [10, 18].”
Learning of grasp adaptation through experience and tactile sensing (2014, methodology): “Since in the controller only the rest length is required to determine the grasp configuration, we move the current grasp configuration towards the closest center of the Gaussian components, which is chosen according to the same distance metric defined above (4).”
Learning Dexterous In-Hand Manipulation. (2018, unknown): “tactile sensing [60, 34, 35, 36].”
Hierarchical Fingertip Space: A Unified Framework for Grasp Planning and In-Hand Grasp Adaptation (2016, background): “Hybrid position and force control has also been studied [39]–[42] as well as impedance control [43]–[46].”
Recent Advances in Robot Learning from Demonstration (2020, unknown): “A popular category of policy learning methods assume direct access to the state and utilize it as the input to the policy (15, 16, 26, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84).”
Constraints extraction from asymmetrical bimanual tasks and their use in coordinated behavior (2018, methodology): “a demonstrator needs to use both hands for unscrewing a light bulb, using a 16 DOF Allegro hand on a stationary robotic arm [13]).”
Variable Impedance Control in End-Effector Space: An Action Space for Reinforcement Learning in Contact-Rich Tasks (2019, background): “Similar use of variable impedance as an action representation for LfD has been demonstrated to be successful for adaptive grasping [26], manipulation of deformable objects [22], and co-adaptation to human workers [36].”
A Review of Tactile Information: Perception and Action Through Touch (2020, methodology): “Some researchers extracted the contact-level information [90], [100] and object-level information [98], [114], [120], [137] using analytical methods, while others computed them with data-driven approaches [184]–[187].”