Tendon-Driven Variable Impedance Control Using Reinforcement Learning

Malhotra, Mark; Theodorou, Evangelos A.; Matsuoka, Yoky; Todorov, Emo

doi:10.15607/rss.2012.viii.047

Cited by 11 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Adaptability: passive adaptation to environmental changes and safe human-robot interaction with mechanical compliance [1,4,5] CONTACT Kuniyuki Takahashi takahashi@sugano.mech.waseda.ac.jp Supplemental data for this article can be accessed https://doi.org/10.1080/01691864.2017.1383939.…”

Section: Background and Advantages Of Robots With Joint Flexibilitymentioning

confidence: 99%

“…When the modeling of the robot and the environment is difficult, model-free machine-learning can be an attractive approach for motion generation with adaptation to dynamic and uncertain environments. However, one of the difficulties of reinforcement learning is that the number of trials would grow with the increase of the dimensionality of the state space [5,11].…”

Section: Related Research On Flexible Joint Robotsmentioning

confidence: 99%

“…During the training of an RNN, the forward calculation in (1) and 2, and backward calculation in (4), (5), and (6) are performed alternately. The criterion of minimum training iterations is as follows: in the case of crankturning, the robot successfully rotates the crank 10 times repeatedly, and in the case of door-opening/closing, the robot successfully opens and closes the door 30 times repeatedly covering the specified range of the door angles (start: 8 • , end: 138 • ).…”

Section: Effectiveness Of Motor Babbling With Active-passive Motionsmentioning

confidence: 99%

See 2 more Smart Citations

Dynamic motion learning for multi-DOF flexible-joint robots using active–passive motor babbling through deep learning

et al. 2017

View full text Add to dashboard Cite

This paper proposes a learning strategy for robots with flexible joints having multi-degrees of freedom in order to achieve dynamic motion tasks. In spite of there being several potential benefits of flexible-joint robots such as exploitation of intrinsic dynamics and passive adaptation to environmental changes with mechanical compliance, controlling such robots is challenging because of increased complexity of their dynamics. To achieve dynamic movements, we introduce a twophase learning framework of the body dynamics of the robot using a recurrent neural network motivated by a deep learning strategy. The proposed methodology comprises a pre-training phase with motor babbling and a fine-tuning phase with additional learning of the target tasks. In the pretraining phase, we consider active and passive exploratory motions for efficient acquisition of body dynamics. In the fine-tuning phase, the learned body dynamics are adjusted for specific tasks. We demonstrate the effectiveness of the proposed methodology in achieving dynamic tasks involving constrained movement requiring interactions with the environment on a simulated robot model and an actual PR2 robot both of which have a compliantly actuated seven degree-of-freedom arm. The results illustrate a reduction in the required number of training iterations for task learning and generalization capabilities for untrained situations.

show abstract

Section: Background and Advantages Of Robots With Joint Flexibilitymentioning

confidence: 99%

Section: Related Research On Flexible Joint Robotsmentioning

confidence: 99%

Section: Effectiveness Of Motor Babbling With Active-passive Motionsmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic motion learning for multi-DOF flexible-joint robots using active–passive motor babbling through deep learning

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Force control has long been recognized as a powerful tool for executing complex robotic motion skills [5]. A number of previous works have addressed learning variableimpedance control policies using reinforcement learning, which uses a reward function and trial-and-error [6], [7], [8]. Although such methods are highly automated, they typically require extensive system interaction during learning, and address simpler behaviors.…”

Section: Related Workmentioning

confidence: 99%

Learning force-based manipulation of deformable objects from multiple demonstrations

Lee

Gupta

et al. 2015

2015 IEEE International Conference on Robotics and Automation (ICRA)

111

View full text Add to dashboard Cite

Manipulation of deformable objects often requires a robot to apply specific forces to bring the object into the desired configuration. For instance, tightening a knot requires pulling on the ends, flattening an article of clothing requires smoothing out wrinkles, and erasing a whiteboard requires applying downward pressure. We present a method for learning force-based manipulation skills from demonstrations. Our approach uses non-rigid registration to compute a warping function that transforms both the end-effector poses and forces in each demonstration into the current scene, based on the configuration of the object. Our method then uses the variation between the demonstrations to extract a single trajectory, along with time-varying feedback gains that determine how much to match poses or forces. This results in a learned variableimpedance control strategy that trades off force and position errors, providing for the right level of compliance that applies the necessary forces at each stage of the motion. We evaluate our approach by tying knots in rope, flattening towels, and erasing a whiteboard.

show abstract

“…It realized simultaneous regulation of motion trajectory and impedance gains using DMPs. This algorithm has been successfully extended to high-dimensional robotic tasks such as opening door, picking up pens [ 34 ], box flipping task [ 35 ] and sliding switch task for tendon-driven hand [ 36 ]. Stulp [ 29 ] further studied the applicability of PI 2 in stochastic force field, and it was able to find motor policies that qualitatively replicate human movement.…”

Section: Introductionmentioning

confidence: 99%

Efficient Force Control Learning System for Industrial Robots Based on Variable Impedance Control

Zhang

Xia

et al. 2018

Sensors

View full text Add to dashboard Cite

Learning variable impedance control is a powerful method to improve the performance of force control. However, current methods typically require too many interactions to achieve good performance. Data-inefficiency has limited these methods to learn force-sensitive tasks in real systems. In order to improve the sampling efficiency and decrease the required interactions during the learning process, this paper develops a data-efficient learning variable impedance control method that enables the industrial robots automatically learn to control the contact force in the unstructured environment. To this end, a Gaussian process model is learned as a faithful proxy of the system, which is then used to predict long-term state evolution for internal simulation, allowing for efficient strategy updates. The effects of model bias are reduced effectively by incorporating model uncertainty into long-term planning. Then the impedance profiles are regulated online according to the learned humanlike impedance strategy. In this way, the flexibility and adaptivity of the system could be enhanced. Both simulated and experimental tests have been performed on an industrial manipulator to verify the performance of the proposed method.

show abstract

Tendon-Driven Variable Impedance Control Using Reinforcement Learning

Cited by 11 publications

References 24 publications

Dynamic motion learning for multi-DOF flexible-joint robots using active–passive motor babbling through deep learning

Dynamic motion learning for multi-DOF flexible-joint robots using active–passive motor babbling through deep learning

Learning force-based manipulation of deformable objects from multiple demonstrations

Efficient Force Control Learning System for Industrial Robots Based on Variable Impedance Control

Contact Info

Product

Resources

About