Robust control of a powered transfemoral prosthesis device with experimental verification

Azimi, Vahid; Shu, Tony; Zhao, Huihua; Ambrose, Eric; Ames, Aaron D.; Simon, Dan

doi:10.23919/acc.2017.7963005

Cited by 17 publications

(10 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system is modeled as a 5-link planar bipedal robot with configuration space Q R : θ = (θ sf , θ sk , θ sh , θ nsh , θ nsk ), for the joint angles of the stance foot, stance knee, stance hip, non-stance hip, and non-stance knee, respectively, as shown in Fig. 1 [23].…”

Section: A Ampro With Pd Controllermentioning

confidence: 99%

Recurrent Neural Network Control of a Hybrid Dynamical Transfemoral Prosthesis with EdgeDRNN Accelerator

Gao

Gehlhar

Ames

et al. 2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

Lower leg prostheses could improve the life quality of amputees by increasing comfort and reducing energy to locomote, but currently control methods are limited in modulating behaviors based upon the human's experience. This paper describes the first steps toward learning complex controllers for dynamical robotic assistive devices. We provide the first example of behavioral cloning to control a powered transfemoral prostheses using a Gated Recurrent Unit (GRU) based recurrent neural network (RNN) running on a custom hardware accelerator that exploits temporal sparsity. The RNN is trained on data collected from the original prosthesis controller. The RNN inference is realized by a novel EdgeDRNN accelerator in real-time. Experimental results show that the RNN can replace the nominal PD controller to realize endto-end control of the AMPRO3 prosthetic leg walking on flat ground and unforeseen slopes with comparable tracking accuracy. EdgeDRNN computes the RNN about 240 times faster than real time, opening the possibility of running larger networks for more complex tasks in the future. Implementing an RNN on this real-time dynamical system with impacts sets the ground work to incorporate other learned elements of the human-prosthesis system into prosthesis control.

show abstract

Section: A Ampro With Pd Controllermentioning

confidence: 99%

Recurrent Neural Network Control of a Hybrid Dynamical Transfemoral Prosthesis with EdgeDRNN Accelerator

Gao

Gehlhar

Ames

et al. 2020

2020 IEEE International Conference on Robotics and Automation (ICRA)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The idea of impedance control is to regulate both position and force by specifying a dynamic relationship between them, 25,33 which is appropriate for the control problems associated with human-robot interactions. [34][35][36][37][38][39][40][41] The reference impedance model chosen in this work is a second-order dynamics given by: (11) where I d , C d , and K d are the desired inertia, damping, and stiffness of the reference model, respectively. ϕ d ,φ d , andφ d are the initially designed trajectory, desired velocity, and the desired acceleration of ankle prosthesis, respectively.…”

Section: Dynamics Of the Reference Impedance Modelmentioning

confidence: 99%

A Novel Robust Model Reference Adaptive Impedance Control Scheme for an Active Transtibial Prosthesis

Heidarzadeh¹,

Sharifi²,

Salarieh³

et al. 2019

Robotica

View full text Add to dashboard Cite

SummaryIn this paper, a novel robust model reference adaptive impedance control (RMRAIC) scheme is presented for an active transtibial ankle prosthesis. The controller makes the closed loop dynamics of the prosthesis similar to a reference impedance model and provides asymptotic tracking of the response trajectory of this impedance model. The interactions between human and prosthesis are taken into account by designing a second-order reference impedance model. The proposed controller is robust against parametric uncertainties in the nonlinear dynamic model of the prosthesis. Also, the controller has robustness against bounded uncertainties due to unavailable ground reaction forces and unmeasurable feedbacks of accelerations at the socket place. Moreover, an appropriate Series Elastic Actuator (SEA) mechanism for the prosthetic ankle is included in this work and its effects are discussed. Tracking performance and stability of the closed-loop system are proven via the Lyapunov stability analysis. Using simulations on an overall amputee prosthetic foot system, the effectiveness of the proposed RMRAIC controller is investigated for the task of level ground walking.

show abstract

“…Most of these above‐knee prostheses use the finite state impedance control [2] and variable impedance control strategy [3]. There are several limitations of impedance control method such as problems in obtaining sub‐phases, no feedback, exhaustive tuning of parameters [3]. Therefore, a simple control technique and robust design approach are the pre‐requisite objectives for control engineers to develop a cost‐effective above‐knee prosthesis.…”

Section: Introductionmentioning

confidence: 99%

“…To avoid buckling and to provide stability during initial stance, powered knee prosthetic leg with motor or damper actuated have been developed in the past. Most of these above-knee prostheses use the finite state impedance control [2] and variable impedance control strategy [3]. There are several limitations of impedance control method such as problems in obtaining sub-phases, no feedback, exhaustive tuning of parameters [3].…”

mentioning

confidence: 99%

User‐feedback based robust and simplified damping control for affordable transfemoral prosthesis

et al. 2020

View full text Add to dashboard Cite

The high cost of the powered prosthetic leg is one of the challenges for the lower-limb amputees in developing countries. This Letter presents a robust design and simplified control strategy for a cost-effective magnetorheological-damper-based prosthetic leg for transfemoral amputees. Herein, a simple sensing technique having an accelerometer and two force-sensing resistors is incorporated to obtain the gait events. Based on gait phase detection, a simplified if-else rule-based control strategy is implemented. A comparison between angles of prosthetic and intact knee is presented to demonstrate the effectiveness of the proposed design.

show abstract

Robust control of a powered transfemoral prosthesis device with experimental verification

Cited by 17 publications

References 19 publications

Recurrent Neural Network Control of a Hybrid Dynamical Transfemoral Prosthesis with EdgeDRNN Accelerator

Recurrent Neural Network Control of a Hybrid Dynamical Transfemoral Prosthesis with EdgeDRNN Accelerator

A Novel Robust Model Reference Adaptive Impedance Control Scheme for an Active Transtibial Prosthesis

User‐feedback based robust and simplified damping control for affordable transfemoral prosthesis

Contact Info

Product

Resources

About