Stochastic Estimation of Multi-Variable Human Ankle Mechanical Impedance

Rastgaar, Mo; Ho, Patrick; Lee, Hyunglae; Krebs, Hermano Igo; Hogan, Neville

doi:10.1115/dscc2009-2643

Cited by 22 publications

(37 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(6) maintained the ankle close to its natural position, preventing drift, However, Eq. (5) uses the torque measurement of the total system, including the torque provided by the output dynamics of the Anklebot, the torque caused by the acceleration of the inertia of the shoe, and the torque due to the impedance of the ankle (Rastgaar et al 2009). Using this measurement would result in the impedance estimation of the whole system.…”

Section: Target Matrixmentioning

confidence: 99%

Estimating the multivariable human ankle impedance in dorsi-plantarflexion and inversion-eversion directions using EMG signals and artificial neural networks

Dallali

Knop

Castelino

et al. 2017

Int J Intell Robot Appl

View full text Add to dashboard Cite

The use of a suitably designed ankle-foot prosthesis is essential for transtibial amputees to regain lost mobility. A desired ankle-foot prosthesis must be able to replicate the function of a healthy human ankle by transferring the ground reaction forces to the body, absorbing shock during contact, and providing propulsion. During the swing phase of walking, the human ankle is soft and relaxed; however, it hardens as it bears the body weight and provides force for push-off. The stiffness is one of the components of the mechanical impedance, and it varies with muscle activation (Stochastic estimation of human ankle mechanical impedance in medial-lateral direction, 2014, Stochastic estimation of the multivariable mechanical impedance of the human ankle with active muscles, 2010). This study defines the relationship between ankle impedance and the lower extremity muscle activations using artificial neural networks (ANN). We used the Anklebot, a highly backdrivable, safe, and therapeutic robot to apply stochastic position perturbations to the human ankle in the sagittal and frontal planes. A previously proposed system identification method was used to estimate the target ankle impedance to train the ANN. The ankle impedance was estimated with relaxed muscles and with lower leg muscle activations at 10 and 20% of the maximum voluntary contraction (MVC) of each individual subject. Given the root mean squared (rms) of the electromyography (EMG) signals, the proposed ANN effectively predicted the ankle impedance with mean accuracy of 89.8 ± 6.1% in DP and mean accuracy of 88.3 ± 5.7% in IE, averaged across three muscle activation levels and all subjects.

show abstract

Section: Target Matrixmentioning

confidence: 99%

Estimating the multivariable human ankle impedance in dorsi-plantarflexion and inversion-eversion directions using EMG signals and artificial neural networks

Dallali

Knop

Castelino

et al. 2017

Int J Intell Robot Appl

View full text Add to dashboard Cite

show abstract

“…The results of the ankle mechanical impedance estimation, using the recently proposed stochastic identification method (Rastgaar et al 2009), is presented in this section. The recorded EMG signals and the estimated ankle impedance were used as input and target data to train the ANN network.…”

Section: Resultsmentioning

confidence: 99%

“…5 and 6 use the torque measurement of the total system. This includes the torque provided by the Anklebot, shoe, and the ankle, as described in (Rastgaar et al 2009). The motion of the Anklebot, shoe, and ankle are the same; therefore, the impedance is in parallel and the Anklebot and shoe can be subtracted from the total estimated impedance.…”

Section: Design Of Target Matrixmentioning

confidence: 99%

“…The advantage of stochastic methods over steady-state procedures is that they provide a quantitative estimate without requiring any a priori assumption about the order or dynamic structure of mechanical impedance (Rastgaar et al 2009). The estimated real and imaginary coefficients of the impedance transfer functions were used as the target matrix for ANN modeling.…”

Section: Design Of Target Matrixmentioning

confidence: 99%

See 1 more Smart Citation

Using lower extremity muscle activity to obtain human ankle impedance in the external–internal direction

Dallali

Knop

Castelino

et al. 2017

Int J Intell Robot Appl

View full text Add to dashboard Cite

The human ankle has a critical role in locomotion and estimating its impedance is essential for human gait rehabilitation. The ankle is the first major joint that regulates the contact forces between the human body and the environment, absorbing shocks during the stance, and providing propulsion during walking. Its impedance varies with the level of the muscle activation. Hence, characterizing the complex relation between the ankle impedance and the lower leg's muscle activation levels may improve our understanding of the neuromuscular characteristics of the ankle. Most ankle-foot prostheses do not have a degree of freedom in the transverse plane, which can cause high amounts of shear stress to be applied to the socket and can lead to secondary injuries. Quantifying the ankle impedance in the transverse plane can guide the design for a variable impedance ankle-foot prosthesis that can significantly reduce the shear stress on the socket. This paper presents the results of applying artificial neural networks (ANN) to learn and estimate the relation between the ankle impedance in the transverse plane under non-load bearing condition using electromyography signals (EMG) from the lower leg muscles. The Anklebot was used to apply pseudorandom perturbations to the human ankle in the transverse plane while the other degrees of freedom (DOF) in the sagittal and frontal planes were constrained. The mechanical impedance of the ankle was estimated using a previously proposed stochastic identification method that describes the ankle impedance as a function of the applied disturbances torques and the ankle motion output. The ankle impedance with relaxed muscles and with the lower leg's muscle activations at 10 and 20% of the maximum voluntary contraction were estimated. The proposed ANN effectively predicts the ankle impedance within 85% accuracy (±5 Nm/rad absolute) for nine out of ten subjects given the root-mean-squared (rms) of the EMG signals. The main contribution of this paper is to quantify the relationship between lower leg muscle EMG signals and the ankle impedance in the transverse plane to pave the way towards designing and controlling this degree of freedom in a future ankle-foot prosthesis.

show abstract

“…The experiment was similar to the tests performed in [26,27] where the Anklebot, a multi-axis lower extremity therapeutic robot, was employed to apply pseudo-random torque perturbations to the ankle in DP and IE directions. The prototype mechanism attached to the Anklebot is shown in Figure 9, where the Anklebot and the ankle-foot mechanism were attached mechanically to each other.…”

Section: Impedance Testmentioning

confidence: 99%

A two-axis cable-driven ankle-foot mechanism

2014

View full text Add to dashboard Cite

This paper describes a novel cable-driven ankle-foot mechanism with two controllable degrees of freedom (DOF) in dorsiflexion-plantarflexion (DP) and inversion-eversion (IE). The presented mechanism is a proof of concept to demonstrate feasibility. Ankle kinematic measurements demonstrate that ankle IE rotations during a step turn are significantly different from walking on a straight path. This suggests that the ankle-foot mechanisms used in prostheses, exoskeletons, and bipedal robots can be improved by controlling a second degree of freedom in the frontal plane. The proposed prototype mechanism is described in detail, and its design considerations and parameters are presented. The mechanism is capable of producing trajectories similar to the human ankle during a step turn. The device shows passive mechanical impedance close to the human ankle mechanical impedance, allowing its mechanical impedance to be controlled using an impedance controller. The presented mechanism is capable of providing key mechanical characteristics similar to the human ankle, including power, range of motion, and weight, suggesting the feasibility of this design concept.

show abstract

Stochastic Estimation of Multi-Variable Human Ankle Mechanical Impedance

Cited by 22 publications

References 4 publications

Estimating the multivariable human ankle impedance in dorsi-plantarflexion and inversion-eversion directions using EMG signals and artificial neural networks

Estimating the multivariable human ankle impedance in dorsi-plantarflexion and inversion-eversion directions using EMG signals and artificial neural networks

Using lower extremity muscle activity to obtain human ankle impedance in the external–internal direction

A two-axis cable-driven ankle-foot mechanism

Contact Info

Product

Resources

About