scholarly journals Dependent-Gaussian-Process-Based Learning of Joint Torques Using Wearable Smart Shoes for Exoskeleton

Sensors ◽  
2020 ◽  
Vol 20 (13) ◽  
pp. 3685
Author(s):  
Jiantao Yang ◽  
Yuehong Yin
Keyword(s):  
Gaussian Process ◽  
Learning Algorithm ◽  
Low Cost ◽  
Joint Kinematics ◽  
Kernel Functions ◽  
Lower Limbs ◽  
Joint Torque ◽  
Human Gait ◽  
Joint Torques ◽  
Composite Kernel

Estimating the joint torques of lower limbs in human gait is a highly challenging task and of great significance in developing high-level controllers for lower-limb exoskeletons. This paper presents a dependent Gaussian process (DGP)-based learning algorithm for joint-torque estimations with measurements from wearable smart shoes. The DGP was established to perform data fusion, and serves as the mathematical foundation to explore the correlations between joint kinematics and joint torques that are embedded deeply in the data. As joint kinematics are used in the training phase rather than the prediction process, the DGP model can realize accurate predictions in outdoor activities by using only the smart shoe, which is low-cost, nonintrusive for human gait, and comfortable to wearers. The design methodology of dynamic specific kernel functions is presented in accordance to prior knowledge of the measured signals. The designed composite kernel functions can be used to model multiple features at different scales, and cope with the temporal evolution of human gait. The statistical nature of the proposed DGP model and the composite kernel functions offer superior flexibility for time-varying gait-pattern learning, and enable accurate joint-torque estimations. Experiments were conducted with five subjects, whose results showed that it is possible to estimate joint torques under different trained and untrained speed levels. Comparisons were made between the proposed DGP and Gaussian process (GP) models. Obvious improvements were achieved when all DGP r2 values were higher than those of GP.

scholarly journals Prediction of Lower Limb Kinetics and Kinematics during Walking by a Single IMU on the Lower Back Using Machine Learning

Sensors ◽  
2019 ◽  
Vol 20 (1) ◽  
pp. 130 ◽  
Author(s):  
Lim ◽  
Kim ◽  
Park
Keyword(s):  
Machine Learning ◽  
Learning Algorithm ◽  
Center Of Mass ◽  
Lower Limbs ◽  
Joint Kinetics ◽  
Joint Torques ◽  
Dynamic Relationship ◽  
Reaction Forces ◽  
Artificial Neural Network Ann ◽  
Parameter Values

Recent studies have reported the application of artificial neural network (ANN) techniques on data of inertial measurement units (IMUs) to predict ground reaction forces (GRFs), which could serve as quantitative indicators of sports performance or rehabilitation. The number of IMUs and their measurement locations are often determined heuristically, and the rationale underlying the selection of these parameter values is not discussed. Using the dynamic relationship between the center of mass (CoM), the GRFs and joint kinetics, we propose the CoM as a single measurement location with which to predict the dynamic data of the lower limbs, using an ANN. Data from seven subjects walking on a treadmill at various speeds were collected from a single IMU worn near the sacrum. The data was segmented by step and numerically processed for integration. Six segment angles of the stance and swing leg, three joint torques, and two GRFs were estimated from the kinematics of the CoM measured from a single IMU sensor, with fair accuracy. These results indicate the importance of the CoM as a dynamic determinant of multi-segment kinetics during walking. The tradeoff between data quantity and wearable convenience can be solved by utilizing a machine learning algorithm based on the dynamic characteristics of human walking.

Sitsen: Passive sitting posture sensing based on wireless devices

2021 ◽  
Vol 17 (7) ◽  
pp. 155014772110248
Author(s):  
Miaoyu Li ◽  
Zhuohan Jiang ◽  
Yutong Liu ◽  
Shuheng Chen ◽  
Marcin Wozniak ◽  
...  
Keyword(s):  
Radio Frequency ◽  
Radio Frequency Identification ◽  
High Performance ◽  
Learning Algorithm ◽  
Low Cost ◽  
Recognition System ◽  
Sitting Posture ◽  
Average Accuracy ◽  
Phase Variations ◽  
Window Approach

Physical health diseases caused by wrong sitting postures are becoming increasingly serious and widespread, especially for sedentary students and workers. Existing video-based approaches and sensor-based approaches can achieve high accuracy, while they have limitations like breaching privacy and relying on specific sensor devices. In this work, we propose Sitsen, a non-contact wireless-based sitting posture recognition system, just using radio frequency signals alone, which neither compromises the privacy nor requires using various specific sensors. We demonstrate that Sitsen can successfully recognize five habitual sitting postures with just one lightweight and low-cost radio frequency identification tag. The intuition is that different postures induce different phase variations. Due to the received phase readings are corrupted by the environmental noise and hardware imperfection, we employ series of signal processing schemes to obtain clean phase readings. Using the sliding window approach to extract effective features of the measured phase sequences and employing an appropriate machine learning algorithm, Sitsen can achieve robust and high performance. Extensive experiments are conducted in an office with 10 volunteers. The result shows that our system can recognize different sitting postures with an average accuracy of 97.02%.

scholarly journals Upper-Limb Isometric Force Feasible Set: Evaluation of Joint Torque-Based Models

Biomechanics ◽  
2021 ◽  
Vol 1 (1) ◽  
pp. 102-117
Author(s):  
Nasser Rezzoug ◽  
Vincent Hernandez ◽  
Philippe Gorce
Keyword(s):  
Upper Limb ◽  
Isometric Force ◽  
Joint Torque ◽  
Force Distribution ◽  
Prediction Errors ◽  
Feasible Set ◽  
Joint Torques ◽  
Motor Abilities ◽  
Capacity Evaluation ◽  
Measured Force

A force capacity evaluation for a given posture may provide better understanding of human motor abilities for applications in sport sciences, rehabilitation and ergonomics. From data on posture and maximum isometric joint torques, the upper-limb force feasible set of the hand was predicted by four models called force ellipsoid, scaled force ellipsoid, force polytope and scaled force polytope, which were compared with a measured force polytope. The volume, shape and force prediction errors were assessed. The scaled ellipsoid underestimated the maximal mean force, and the scaled polytope overestimated it. The scaled force ellipsoid underestimated the volume of the measured force distribution, whereas that of the scaled polytope was not significantly different from the measured distribution but exhibited larger variability. All the models characterized well the elongated shape of the measured force distribution. The angles between the main axes of the modelled ellipsoids and polytopes and that of the measured polytope were compared. The values ranged from 7.3° to 14.3°. Over the entire surface of the force ellipsoid, 39.7% of the points had prediction errors less than 50 N; 33.6% had errors between 50 and 100 N; and 26.8% had errors greater than 100N. For the force polytope, the percentages were 56.2%, 28.3% and 15.4%, respectively.

scholarly journals Deep Convolutional and LSTM Networks on Multi-Channel Time Series Data for Gait Phase Recognition

Sensors ◽  
2021 ◽  
Vol 21 (3) ◽  
pp. 789
Author(s):  
David Kreuzer ◽  
Michael Munz
Keyword(s):  
Gait Analysis ◽  
Time Series Data ◽  
Low Cost ◽  
Analysis Data ◽  
Gait Disorders ◽  
Human Gait ◽  
Series Data ◽  
Ageing Society ◽  
Time Motion ◽  
Gait Phase

With an ageing society comes the increased prevalence of gait disorders. The restriction of mobility leads to a considerable reduction in the quality of life, because associated falls increase morbidity and mortality. Consideration of gait analysis data often alters surgical recommendations. For that reason, the early and systematic diagnostic treatment of gait disorders can spare a lot of suffering. As modern gait analysis systems are, in most cases, still very costly, many patients are not privileged enough to have access to comparable therapies. Low-cost systems such as inertial measurement units (IMUs) still pose major challenges, but offer possibilities for automatic real-time motion analysis. In this paper, we present a new approach to reliably detect human gait phases, using IMUs and machine learning methods. This approach should form the foundation of a new medical device to be used for gait analysis. A model is presented combining deep 2D-convolutional and LSTM networks to perform a classification task; it predicts the current gait phase with an accuracy of over 92% on an unseen subject, differentiating between five different phases. In the course of the paper, different approaches to optimize the performance of the model are presented and evaluated.

scholarly journals Research on Task-Oriented Computation Offloading Decision in Space-Air-Ground Integrated Network

2021 ◽  
Vol 13 (5) ◽  
pp. 128
Author(s):  
Jun Liu ◽  
Xiaohui Lian ◽  
Chang Liu
Keyword(s):  
Energy Consumption ◽  
Decision Process ◽  
Learning Algorithm ◽  
Low Cost ◽  
Transition Process ◽  
Computation Offloading ◽  
Processing Efficiency ◽  
Integrated Network ◽  
Decision Scheme ◽  
Task Offloading

In Space–Air–Ground Integrated Networks (SAGIN), computation offloading technology is a new way to improve the processing efficiency of node tasks and improve the limitation of computing storage resources. To solve the problem of large delay and energy consumption cost of task computation offloading, which caused by the complex and variable network offloading environment and a large amount of offloading tasks, a computation offloading decision scheme based on Markov and Deep Q Networks (DQN) is proposed. First, we select the optimal offloading network based on the characteristics of the movement of the task offloading process in the network. Then, the task offloading process is transformed into a Markov state transition process to build a model of the computational offloading decision process. Finally, the delay and energy consumption weights are introduced into the DQN algorithm to update the computation offloading decision process, and the optimal offloading decision under the low cost is achieved according to the task attributes. The simulation results show that compared with the traditional Lyapunov-based offloading decision scheme and the classical Q-learning algorithm, the delay and energy consumption are respectively reduced by 68.33% and 11.21%, under equal weights when the offloading task volume exceeds 500 Mbit. Moreover, compared with offloading to edge nodes or backbone nodes of the network alone, the proposed mixed offloading model can satisfy more than 100 task requests with low energy consumption and low delay. It can be seen that the computation offloading decision proposed in this paper can effectively reduce the delay and energy consumption during the task computation offloading in the Space–Air–Ground Integrated Network environment, and can select the optimal offloading sites to execute the tasks according to the characteristics of the task itself.

Kinematic Coordination in Human Gait: Relation to Mechanical Energy Cost

1998 ◽  
Vol 79 (4) ◽  
pp. 2155-2170 ◽  
Author(s):  
L. Bianchi ◽  
D. Angelini ◽  
G. P. Orani ◽  
F. Lacquaniti
Keyword(s):  
Energy Expenditure ◽  
Energy Cost ◽  
Time Course ◽  
Sagittal Plane ◽  
Dimensional Space ◽  
Mechanical Energy ◽  
Direction Cosine ◽  
Lower Limbs ◽  
Human Gait ◽  
Plane Orientation

Bianchi, L., D. Angelini, G. P. Orani, and F. Lacquaniti. Kinematic coordination in human gait: relation to mechanical energy cost. J. Neurophysiol. 79: 2155–2170, 1998. Twenty-four subjects walked at different, freely chosen speeds ( V) ranging from 0.4 to 2.6 m s−1, while the motion and the ground reaction forces were recorded in three-dimensional space. We considered the time course of the changes of the angles of elevation of the trunk, pelvis, thigh, shank, and foot in the sagittal plane. These angles specify the orientation of each segment with respect to the vertical and to the direction of forward progression. The changes of the trunk and pelvis angles are of limited amplitude and reflect the dynamics of both right and left lower limbs. The changes of the thigh, shank, and foot elevation are ample, and they are coupled tightly among each other. When these angles are plotted one versus the others, they describe regular loops constrained on a plane. The plane of angular covariation rotates, slightly but systematically, along the long axis of the gait loop with increasing V. The rotation, quantified by the change of the direction cosine of the normal to the plane with the thigh axis ( u 3 t ), is related to a progressive phase shift between the foot elevation and the shank elevation with increasing V. As a next step in the analysis, we computed the mass-specific mean absolute power ( P u ) to obtain a global estimate of the rate at which mechanical work is performed during the gait cycle. When plotted on logarithmic coordinates, P u increases linearly with V. The slope of this relationship varies considerably across subjects, spanning a threefold range. We found that, at any given V > 1 m s−1, the value of the plane orientation ( u 3 t ) is correlated with the corresponding value of the net mechanical power ( P u ). On the average, the progressive rotation of the plane with increasing V is associated with a reduction of the increment of P u that would occur if u 3 t remained constant at the value characteristic of low V. The specific orientation of the plane at any given speed is not the same in all subjects, but there is an orderly shift of the plane orientation that correlates with the net power expended by each subject. In general, smaller values of u 3 t tend to be associated with smaller values of P u and vice versa. We conclude that the parametric tuning of the plane of angular covariation is a reliable predictor of the mechanical energy expenditure of each subject and could be used by the nervous system for limiting the overall energy expenditure.

Multi-Objective Optimization of Human Gait With a Discomfort Function

2016 ◽  
Author(s):  
Hyun-Jung Kwon ◽  
Hyun-Joon Chung ◽  
Yujiang Xiang
Keyword(s):  
Joint Torque ◽  
Upper Body ◽  
Whole Body ◽  
Human Gait ◽  
Multi Objective Optimization ◽  
Joint Angles ◽  
Backward Walking ◽  
Multi Objective ◽  
Body Model ◽  
Neutral Angle

The objective of this study was to develop a discomfort function for including a high DOF upper body model during walking. A multi-objective optimization (MOO) method was formulated by minimizing dynamic effort and the discomfort function simultaneously. The discomfort function is defined as the sum of the squares of deviation of joint angles from their neutral angle positions. The dynamic effort is the sum of the joint torque squared. To investigate the efficacy of the proposed MOO method, backward walking simulation was conducted. By minimizing both dynamic effort and the discomfort function, a 3D whole body model with a high DOF upper body for walking was demonstrated successfully.

scholarly journals Joint torque and velocity optimization for a redundant leg of quadruped robot

2017 ◽  
Vol 14 (5) ◽  
pp. 172988141773189 ◽  
Author(s):  
Taihui Zhang ◽  
Honglei An ◽  
Hongxu Ma
Keyword(s):  
Angular Velocity ◽  
Sagittal Plane ◽  
Load Capacity ◽  
Optimization Criterion ◽  
Quadruped Robot ◽  
Joint Torque ◽  
Quadratic Program ◽  
Hydraulic Actuator ◽  
Joint Torques ◽  
Physical Limitations

Hydraulic actuated quadruped robot similar to BigDog has two primary performance requirements, load capacity and walking speed, so that it is necessary to balance joint torque and joint velocity when designing the dimension of single leg and controlling its motion. On the one hand, because there are three joints per leg on sagittal plane, it is necessary to firstly optimize the distribution of torque and angular velocity of every joint on the basis of their different requirements. On the other hand, because the performance of hydraulic actuator is limited, it is significant to keep the joint torque and angular velocity in actuator physical limitations. Therefore, it is essential to balance the joint torque and angular velocity which have negative correlation under the condition of constant power of the hydraulic actuator. The main purpose of this article is to optimize the distribution of joint torques and velocity of a redundant single leg with joint physical limitations. Firstly, a modified optimization criterion combining joint torques with angular velocity that takes both support phase and flight phase into account is proposed, and then the modified optimization criterion is converted into a normal quadratic programming problem. A kind of recurrent neural network is used to solve the quadratic program problem. This method avoids tremendous matrix inversion and fits for time-varying system. The achieved optimized distribution of joint torques and velocity is useful for aiding mechanical design and the following motion control. Simulation results presented in this article confirm the efficiency of this optimization algorithm.

Physics-Based Seated Posture Prediction for Pregnant Women and Validation Considering Ground and Seat Pan Contacts

2012 ◽  
Vol 134 (7) ◽  
Author(s):  
Bradley Howard ◽  
Aimee Cloutier ◽  
Jingzhou (James) Yang
Keyword(s):  
Pregnant Women ◽  
Objective Function ◽  
Linear Regression Analysis ◽  
Joint Torque ◽  
Joint Torques ◽  
Design Variables ◽  
Optimization Formulation ◽  
Posture Prediction ◽  
Seated Posture ◽  
End Effectors

An understanding of human seated posture is important across many fields of scientific research. Certain demographics, such as pregnant women, have special postural limitations that need to be considered. Physics-based posture prediction is a tool in which seated postures can be quickly and thoroughly analyzed, as long the predicted postures are realistic. This paper proposes and validates an optimization formulation to predict seated posture for pregnant women considering ground and seat pan contacts. For the optimization formulation, the design variables are joint angles (posture); the cost function is dependent on joint torques. Constraints include joint limits, joint torque limits, the distances from the end-effectors to target points, and self-collision avoidance constraints. Three different joint torque cost functions have been investigated to account for the special postural characteristics of pregnant women and consider the support reaction forces (SRFs) associated with seated posture. Postures are predicted for three different reaching tasks in common reaching directions using each of the objective function formulations. The predicted postures are validated against experimental postures obtained using motion capture. A linear regression analysis was used to evaluate the validity of the predicted postures and was the criteria for comparison between the different objective functions. A 56 degree of freedom model was used for the posture prediction. Use of the objective function minimizing the maximum normalized joint torque provided an R2 value of 0.828, proving superior to either of two alternative functions.

Sign in / Sign up

Export Citation Format

Share Document