Calibration of an instrumented treadmill using a precision-controlled device with artificial neural network-based error corrections

Hsieh, Han-Chang; Lin, Hong‐Ping; Lu, Hsuan-Lun; Chen, Ting-Yi; Lu, Tung‐Wu

doi:10.1016/j.gaitpost.2016.01.018

Cited by 7 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Pose was extracted from the video with corresponding foot pressure maps used to train the neural network. Most commonly, ANNs are used as a calibration function which maps raw sensor data from force or pressure sensors into the estimated COP [16][17][18].…”

Section: Introductionmentioning

confidence: 99%

Centre of Pressure Estimation during Walking Using Only Inertial-Measurement Units and End-To-End Statistical Modelling

Podobnik

Kraljić

Zadravec

et al. 2020

Sensors

View full text Add to dashboard Cite

Estimation of the centre of pressure (COP) is an important part of the gait analysis, for example, when evaluating the functional capacity of individuals affected by motor impairment. Inertial measurement units (IMUs) and force sensors are commonly used to measure gait characteristic of healthy and impaired subjects. We present a methodology for estimating the COP solely from raw gyroscope, accelerometer, and magnetometer data from IMUs using statistical modelling. We demonstrate the viability of the method using an example of two models: a linear model and a non-linear Long-Short-Term Memory (LSTM) neural network model. Models were trained on the COP ground truth data measured using an instrumented treadmill and achieved the average intra-subject root mean square (RMS) error between estimated and ground truth COP of 12.3 mm and the average inter-subject RMS error of 23.7 mm which is comparable or better than similar studies so far. We show that the calibration procedure in the instrumented treadmill can be as short as a couple of minutes without the decrease in our model performance. We also show that the magnetic component of the recorded IMU signal, which is most sensitive to environmental changes, can be safely dropped without a significant decrease in model performance. Finally, we show that the number of IMUs can be reduced to five without deterioration in the model performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Centre of Pressure Estimation during Walking Using Only Inertial-Measurement Units and End-To-End Statistical Modelling

Podobnik

Kraljić

Zadravec

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Например, при остановке пользователя полотно дорожки в течение нескольких секунд продолжать движение или сбрасывает скорость, что приводит к негативным последствиям для пользователей, таким как расстройство вестибулярного аппарата, укачивание, потеря равновесия, тошнота [3]. Все эти факторы негативно влияют на погружение человека в виртуальное пространство и препятствуют процессу подготовки [4][5][6][7][8][9].…”

Section: анализ подходов к управлению системами имитации физических нагрузокunclassified

Neural network method of adaptive control system of imitation of physical forces

Obukhov

Дмитриевич²,

Siukhin

et al. 2020

Vestnik of Samara State Technical University. Technical Science

View full text Add to dashboard Cite

This research examines the subject area of physical forces simulation systems, implemented on the basis of controlled running platforms. The time spent by the control system to receive and process information about the state of the user and the system causes a software and hardware delay that prevents the system from responding in a timely manner to the user's natural movement. The control system delay problem cannot be solved using direct data of the states of the man-machine system. The aim of the presented research is to develop a new control method that allows analyzing the state of the user and the platform, forecasting his movements and organizing the process of managing the system for simulating physical forces. The method is implemented using neural networks. The scientific novelty of the method includes in the use of neural networks to solve the problems of forecasting user actions and automating decision-making to control the system for simulating physical forces. Each presented neural network is formed to perform separate tasks. The first is to create a forecast of changes in the states of a man-machine system. The second is to determine whether the forecasted state belongs to any state in the historical data. The third determines the required change in the states of the parameters of the man-machine system to achieve the forecasted state. The possibilities of using the described approach are presented on the example of a treadmill that adapts to the real parameters of the user's locomotion. The results obtained confirm a significant reduction in the complexity of the implementation of the control process after applying the neural network method. The area of application of the neural network control method is adaptive information systems and automatic control systems, in which it is necessary to minimize the system delay time and response to user locomotion.

show abstract

“…In a gait laboratory, each subject was asked to walk barefoot on an 8-meter walkway at his self-selected speed, as well as on an instrumented treadmill which had been calibrated for accurate COP measurements (Hsieh et al, 2016), at the same self-selected speed (average speed of six over-ground gait trials). The subjects were allowed a 6-minute practice session on the treadmill before data collection in order to eliminate any possible effects on gait parameters due to unfamiliarity with treadmill locomotion (Matsas et al, 2000).…”

Section: Data Collectionmentioning

confidence: 99%

Control of the body's centre of mass motion during over-ground and treadmill walking in the pelvic reference frame: Implications for wearable sensing

Huang

Lee

et al. 2017

JBSE

Self Cite

View full text Add to dashboard Cite

Treadmill-based training protocols have played an important role in gait and balance training in rehabilitation settings, and the position of the body's center of mass (COM) relative to the base of support provides an important measurement of postural stability. With the advance in wearable technology, the movement of the COM has been estimated using inertial measurement units (IMU) or by placing a single marker on the pelvis. However, there has not been a consensus over the differences in the movement control of COM relative to pelvis between treadmill walking (TW) and over-ground walking (OW). The current study aimed to quantify and compare the motions of the COM with respect to the pelvic reference frame during TW and OW. Fifteen young male adults walked barefoot on an 8-meter walkway, as well as on an instrumented treadmill at their self-selected speed while their pelvis position, COM positions, and center of pressure (COP) positions were measured. Paired ttests were performed between OW and TW for all the variables. The results showed that compared to OW, treadmill walking produced significantly greater variations of COM displacements in every direction, namely anterior/posterior, proximal/distal, and medial/lateral, and significantly greater excursions of the COM in the medial/lateral direction with respect to the pelvis coordinate system. The results suggest that the estimation of the COM position with an IMU attached to the pelvis is more accurate during OW than TW.

show abstract

Calibration of an instrumented treadmill using a precision-controlled device with artificial neural network-based error corrections

Cited by 7 publications

References 25 publications

Centre of Pressure Estimation during Walking Using Only Inertial-Measurement Units and End-To-End Statistical Modelling

Centre of Pressure Estimation during Walking Using Only Inertial-Measurement Units and End-To-End Statistical Modelling

Neural network method of adaptive control system of imitation of physical forces

Control of the body's centre of mass motion during over-ground and treadmill walking in the pelvic reference frame: Implications for wearable sensing

Contact Info

Product

Resources

About