RNN-Based On-Line Continuous Gait Phase Estimation from Shank-Mounted IMUs to Control Ankle Exoskeletons

Seo, Keehong; Park, Young Jin; Lee, Jusuk; Hyung, Seungyong; Lee, Minhyung; Kim, Jeonghun; Choi, Hyundo; Shim, Youngbo

doi:10.1109/icorr.2019.8779554

Cited by 40 publications

(42 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…H UMAN gait phase estimation has been studied for clinical or rehabilitation purpose [1]- [4] and for developing assistive robotic devices, such as powered prostheses [5]- [9] or exoskeletons [10], [11]. For the clinical purpose, a set of vision-based motion capture systems and force plate is conventionally utilized to evaluate the rehabilitation process or to monitor patient's gait abnormalities by observing patient's motion in the well-controlled space [2], [4].…”

Section: Introductionmentioning

confidence: 99%

“…Thanks to rapid advancements in machine learning, gait phase estimation methods have evolved using multiple sensor information. For instance, Seo et al [10] estimated the gait phase for their ankle exoskeleton, relying on a shankmounted IMU. To achieve a real-time estimation for their phase-based control, they proposed to use Recurrent Neural Networks (RNNs) with additional information from the foot pressure sensors during their model training.…”

Section: Introductionmentioning

confidence: 99%

“…Similarly, in [11], researchers also utilized a neural network learning model to estimate the gait phase based on sensor-fusion: the encoders at the hip, IMUs at the thigh and trunk. Regardless of the participants, they both achieved more robust and accurate estimation of a continuous variable with the learning technique [10], [11]. These methods, however, are not applicable to the lowerlimb prostheses due to the difference between the exoskeleton and the prosthesis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Continuous Gait Phase Estimation Using LSTM for Robotic Transfemoral Prosthesis Across Walking Speeds

Lee

Hong

Hur

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

User gait phase estimation plays a key role for the seamless control of the lower-limb robotic assistive devices (e.g., exoskeletons or prostheses) during ambulation. To achieve this, several studies have attempted to estimate the gait phase using a thigh or shank angle. However, their estimation resulted in some deviation from the actual walking and varied across the walking speeds. In this study, we investigated the different setups using for the machine learning approach to obtain more accurate and consistent gait phase estimation for the robotic transfemoral prosthesis over different walking speeds. Considering the transfemoral prosthetic application, we proposed two different sensor setups: i) the angular positions and velocities of both thigh and torso (S1) and ii) the angular positions and velocities of both thigh and torso, and heel force data (S2). The proposed setups and method are experimentally evaluated with three healthy young subjects at four different walking speeds: 0.5, 1.0, 1.5, and 2.0 m/s. Both results showed robust and accurate gait phase estimation with respect to the ground truth (loss value of S1: 4.54e-03 Vs. S2: 4.70e-03). S1 had the advantage of a simple equipment setup using only two IMUs, while S2 had the advantage of estimating more accurate heel-strikes than S1 by using additional heel force data. The choice between the two sensor setups can depend on the researchers' preference in consideration of the device setup or the focus of the interest.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Continuous Gait Phase Estimation Using LSTM for Robotic Transfemoral Prosthesis Across Walking Speeds

Lee

Hong

Hur

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…The results of the study are comparable with the study [14], who have conducted the gait phase estimation based on AOs on healthy people. Compared with the study [11] which adopted the extended Kalman filter to estimate gait phase and study [13] which adopted RNN-Based method to estimate gait phase, our results show better performance. In addition, AOs have little parameters to tune, and the parameters was adaptive for different subjects.…”

Section: Continuous Gait Phase Estimationmentioning

confidence: 65%

“…However, the phase estimation results still need further improvement. Seo et al have conducted online continuous gait phase estimation to control ankle exoskeletons, but the estimation performance also need further improvement [13]. Adaptive oscillators (AOs) have also been used to continuous gait phase estimation on healthy people and exoskeleton [14], [15], and AOs have shown better performance than Kalman filter and RNN methods, since its inherent synchronization properties provided advantages in continuous gait phase estimation.…”

Section: Introductionmentioning

confidence: 99%

Online Estimation of Continuous Gait Phase for Robotic Transtibial Prostheses Based on Adaptive Oscillators

Crea

Vitiello

et al. 2020

2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM)

View full text Add to dashboard Cite

Continuous gait phase plays an important role in wearable robot control. This study focuses on the online estimation of continuous gait phase based on robotic transtibial prosthesis signals. First, we adopt the prosthetic foot deformation information to detect the heel strike as the start timing (reset 0 rad) of one gait cycle. Then we conduct the gait phase estimation based on adaptive oscillators using the prosthetic shank angle signal as input. Three transtibial amputees were recruited in this study and they walked on the treadmill at different speeds (slow, normal and fast) and on different ramps (10 • , 5 • , 0 • , -5 • and -10 • ) in the experiment. The root-meansquare error between online estimation result and ground truth gait phase is calculated. The maximum and minimum errors are 0.147 rad and 0.058 rad, and the corresponding ratios in one gait cycle are 2.34% and 0.92%. This study achieves good performance and provides an effective method to estimate the continuous gait phase, which will instruct robotic transtibial prosthesis control.

show abstract

Research on Environmental Slope Prediction of Knee Assisted Exoskeleton Based on Multi-source Signals

Chen

Zhu

et al. 2022

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

RNN-Based On-Line Continuous Gait Phase Estimation from Shank-Mounted IMUs to Control Ankle Exoskeletons

Cited by 40 publications

References 18 publications

Continuous Gait Phase Estimation Using LSTM for Robotic Transfemoral Prosthesis Across Walking Speeds

Continuous Gait Phase Estimation Using LSTM for Robotic Transfemoral Prosthesis Across Walking Speeds

Online Estimation of Continuous Gait Phase for Robotic Transtibial Prostheses Based on Adaptive Oscillators

Research on Environmental Slope Prediction of Knee Assisted Exoskeleton Based on Multi-source Signals

Contact Info

Product

Resources

About