Transferring Human Impedance Behavior to Heterogeneous Variable Impedance Actuators

Howard, Matthew; Braun, David J.; Vijayakumar, Sethu

doi:10.1109/tro.2013.2256311

Cited by 51 publications

(50 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…muscle-tendon) bandwidth-limitations [56]. Regarding this last point, we note that care should also be taken when transferring experimental observations from humans to robots or vice versa [57]. This is because impedance strategies by humans are subject to inherent bandwidth limitations of the muscle-tendon system and as such, may not contain important task-specific features that could be utilized by robots equipped with fast actuators.…”

Section: ) Simulation and Experimental Resultsmentioning

confidence: 99%

Robots Driven by Compliant Actuators: Optimal Control Under Actuation Constraints

et al. 2013

Self Cite

View full text Add to dashboard Cite

Abstract-Anthropomorphic robots that aim to approach human performance agility and efficiency are typically highly redundant not only in their kinematics but also in actuation. Variableimpedance actuators, used to drive many of these devices, are capable of modulating torque and impedance (stiffness and/or damping) simultaneously, continuously, and independently. These actuators are, however, nonlinear and assert numerous constraints, e.g., range, rate, and effort limits on the dynamics. Finding a control strategy that makes use of the intrinsic dynamics and capacity of compliant actuators for such redundant, nonlinear, and constrained systems is nontrivial. In this study, we propose a framework for optimization of torque and impedance profiles in order to maximize task performance, which is tuned to the complex hardware and incorporating real-world actuation constraints. Simulation study and hardware experiments 1) demonstrate the effects of actuation constraints during impedance control, 2) show applicability of the present framework to simultaneous torque and temporal stiffness optimization under constraints that are imposed by real-world actuators, and 3) validate the benefits of the proposed approach under experimental conditions.

show abstract

Section: ) Simulation and Experimental Resultsmentioning

confidence: 99%

Robots Driven by Compliant Actuators: Optimal Control Under Actuation Constraints

et al. 2013

Self Cite

View full text Add to dashboard Cite

show abstract

“…Inspired to these studies, a learning control technique is proposed in [11] to optimally adapt robot's impedance during the interaction with dynamic environments and humans. Learning techniques have been adopted also to extract and transfer impedance modulation strategies from humans to variable impedance robots [12] or, complementary, to teach variable stiffness tasks to robots through physical interaction with a human operator [13].…”

Section: Related Workmentioning

confidence: 99%

Variable Impedance Control of Redundant Manipulators for Intuitive Human–Robot Physical Interaction

2015

View full text Add to dashboard Cite

This paper presents an experimental study on human-robot comanipulation in the presence of kinematic redundancy. The objective of the work is to enhance the performance during human-robot physical interaction by combining Cartesian impedance modulation and redundancy resolution. Cartesian impedance control is employed to achieve a compliant behavior of the robot's end effector in response to forces exerted by the human operator. Different impedance modulation strategies, which take into account the human's behavior during the interaction, are selected with the support of a simulation study and then experimentally tested on a 7-degree-of-freedom KUKA LWR4. A comparative study to establish the most effective redundancy resolution strategy has been made by evaluating different solutions compatible with the considered task. The experiments have shown that the redundancy, when used to ensure a decoupled apparent inertia at the end effector, allows enlarging the stability region in the impedance parameters space and improving the performance. On the other hand, the variable impedance with a suitable modulation strategy for parameters' tuning outperforms the constant impedance, in the sense that it enhances the comfort perceived by humans during manual guidance and allows reaching a favorable compromise between accuracy and execution time.Index Terms-Force control, physical human-robot interaction, redundant robots. 1552-3098

show abstract

“…The principles observed in human impedance adaptation were transfered to an adaptive impedance controller in [33]. Task specific impedance principles captured via a learned cost-function were transferred from humans to robots in [34], and direct mirroring of human stiffness was proposed for tele-operation applica-tions in [35]. In the field of LfD, it has been proposed to derive stiffness variations via kinematic variability in demonstrated data [36].…”

Section: Learning Varying Stiffness Controlmentioning

confidence: 99%

Learning motions from demonstrations and rewards with time-invariant dynamical systems based policies

Rey

Kronander

Farshidian³

et al. 2017

Auton Robot

View full text Add to dashboard Cite

An important challenge when using Reinforcement Learning for learning motions in robotics is the choice of parameterization for the policy. We use Gaussian Mixture Regression to extract a parameterization with relevant non-linear features from a set of demonstrations of a motion following the paradigm of Learning from Demonstration. The resulting parameterization takes the form of a non-linear time-invariant dynamical system (DS). We use this time-invariant DS as a parameterized policy for a variant of the PI 2 policy search algorithm. This paper contributes by adapting PI 2 for our time-invariant motion representation. We introduce two novel parameter exploration schemes that can be used to 1) sample model parameters to achieve a uniform exploration in state space and 2) explore while ensuring stability of the resulting motion model. Additionally, a state dependent stiffness profile is learned simultaneously to the reference trajectory and both are used together in a variable impedance control architecture. This learning architecture is validated in a hardware experiment consisting of a digging task using a KUKA LWR platform.

show abstract

Transferring Human Impedance Behavior to Heterogeneous Variable Impedance Actuators

Cited by 51 publications

References 58 publications

Robots Driven by Compliant Actuators: Optimal Control Under Actuation Constraints

Robots Driven by Compliant Actuators: Optimal Control Under Actuation Constraints

Variable Impedance Control of Redundant Manipulators for Intuitive Human–Robot Physical Interaction

Learning motions from demonstrations and rewards with time-invariant dynamical systems based policies

Contact Info

Product

Resources

About