Motion Prediction with Artificial Neural Networks Using Wearable Strain Sensors Based on Flexible Thin Graphite Films

2019 ◽  
Vol 826 ◽  
pp. 111-116
Author(s):  
Takahiro Kanokoda ◽  
Yuki Kushitani ◽  
Moe Shimada ◽  
Jun Ichi Shirakashi
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
Wearable Devices ◽  
Human Motion ◽  
Strain Sensors ◽  
Motion Prediction ◽  
Prediction System ◽  
Hand Motion ◽  
Motion Data ◽  
Human Motion Prediction

A human motion prediction system can be used to estimate human gestures in advance to the actual action for reducing delays in interactive system. We have already reported a method of simple and easy fabrication of strain sensors and wearable devices using pyrolytic graphite sheets (PGSs). The wearable electronics could detect various types of human motion, with high durability and fast response. In this study, we have demonstrated hand motion prediction by neural networks (NNs) using hand motion data obtained from data gloves based on PGSs. In our experiments, we measured hand motions of subjects for learning. We created 4-layered NNs to predict human hand motion in real-time. As a result, the proposed system successfully predicted hand motion in real-time. Therefore, these results suggested that human motion prediction system using NNs is able to forecast various types of human behavior using human motion data obtained from wearable devices based on PGSs.

scholarly journals Few-shot Human Motion Prediction via Learning Novel Motion Dynamics

2020 ◽  
Author(s):  
Chuanqi Zang ◽  
Mingtao Pei ◽  
Yu Kong
Keyword(s):  
Temporal Structure ◽  
Human Motion ◽  
Training Data ◽  
Motion Prediction ◽  
Motion Data ◽  
Human Motion Prediction ◽  
Novel Approach ◽  
Spatio Temporal ◽  
Motion Dynamics

Human motion prediction is a task where we anticipate future motion based on past observation. Previous approaches rely on the access to large datasets of skeleton data, and thus are difficult to be generalized to novel motion dynamics with limited training data. In our work, we propose a novel approach named Motion Prediction Network (MoPredNet) for few-short human motion prediction. MoPredNet can be adapted to predicting new motion dynamics using limited data, and it elegantly captures long-term dependency in motion dynamics. Specifically, MoPredNet dynamically selects the most informative poses in the streaming motion data as masked poses. In addition, MoPredNet improves its encoding capability of motion dynamics by adaptively learning spatio-temporal structure from the observed poses and masked poses. We also propose to adapt MoPredNet to novel motion dynamics based on accumulated motion experiences and limited novel motion dynamics data. Experimental results show that our method achieves better performance over state-of-the-art methods in motion prediction.

scholarly journals Acceleration Feature Extraction of Human Body Based on Wearable Devices

Energies ◽  
2021 ◽  
Vol 14 (4) ◽  
pp. 924
Author(s):  
Zhenzhen Huang ◽  
Qiang Niu ◽  
Ilsun You ◽  
Giovanni Pau
Keyword(s):  
Genetic Algorithm ◽  
Feature Extraction ◽  
Kalman Filter ◽  
Human Body ◽  
Wearable Devices ◽  
Human Motion ◽  
Data Set ◽  
Feature Extraction Method ◽  
Motion Data ◽  
Kalman Filter Algorithm

Wearable devices used for human body monitoring has broad applications in smart home, sports, security and other fields. Wearable devices provide an extremely convenient way to collect a large amount of human motion data. In this paper, the human body acceleration feature extraction method based on wearable devices is studied. Firstly, Butterworth filter is used to filter the data. Then, in order to ensure the extracted feature value more accurately, it is necessary to remove the abnormal data in the source. This paper combines Kalman filter algorithm with a genetic algorithm and use the genetic algorithm to code the parameters of the Kalman filter algorithm. We use Standard Deviation (SD), Interval of Peaks (IoP) and Difference between Adjacent Peaks and Troughs (DAPT) to analyze seven kinds of acceleration. At last, SisFall data set, which is a globally available data set for study and experiments, is used for experiments to verify the effectiveness of our method. Based on simulation results, we can conclude that our method can distinguish different activity clearly.

scholarly journals Multitask Non-Autoregressive Model for Human Motion Prediction

2020 ◽  
pp. 1-1
Author(s):  
Bin Li ◽  
Jian Tian ◽  
Zhongfei Zhang ◽  
Hailin Feng ◽  
Xi Li
Keyword(s):  
Autoregressive Model ◽  
Human Motion ◽  
Motion Prediction ◽  
Human Motion Prediction
Sign in / Sign up

Export Citation Format

Share Document