An Adaptive Algorithm to Improve Energy Efficiency in Wearable Activity Recognition Systems

There are several large datasets available that are captured from motion tracking systems which could be useful to train wearable human activity recognition (HAR) systems, if only their spatial data could be mapped into the equivalent inertial measurement unit (IMU) data that would be sensed on the body. In this paper, we describe a mapping from 3D Vicon motion tracking data to data collected from a BlueSense on-body IMU. We characterise the error incurred in order to discern the extent to which it is possible to generate useful training data for a wearable activity recognition system from data collected with a motion capture system. We analyse this by mapping Vicon motion tracking data to rotational velocity and linear acceleration at the head, and compare this to actual gyroscope and accelerometer data collected by an IMU mounted on the head. In a 15 min dataset comprising three static activities—sitting, standing and lying down—we find that 95% of the reconstructed gyroscope data is within an error of [−7.25;+7.46] \(deg \cdot s^{-1}\), while 95% of the reconstructed accelerometer data was contained within [−96.1;+72.9] \(m \cdot G\). However, when we introduce more movement by including data collected while walking this increases to [−19.0;+18.2] \(deg \cdot s^{-1}\) for the gyroscope and [−208;+186] \(m \cdot G\) for the accelerometer. We conclude that generating accurate IMU data from motion capture datasets is possible and could be useful in providing larger volumes of data for activity recognition tasks and in helping enable advanced, data-hungry techniques such as deep learning to be employed on a larger scale within the domain of human activity recognition.

We investigate the use of magnetic field disturbances as features for motion based, wearable activity recognition systems. Such disturbances are mostly caused by large metallic objects and electrical appliances, both of which are often involved in human activities. We propose to detect them by subtracting angular velocity values computed from the changes in the magnetic field vector from gyroscope signals. We argue that for activities that are associated with specific objects or devices such features increase system robustness against motion variations, sensor displacement and inter user differences. On a previously published data set of 8 gym exercises we demonstrate that our approach can improve the recognition by up to 31% over gyroscope only and up to 17% over a combination of a gyroscope and 3D accelerometer. Improvements of 9.5% are also demonstrated for user independent training as well as for the case of displaced sensors.A particularly interesting result is the fact that adding the magnetic disturbance features significantly improves recognition based on the vector norm of accelerometers and gyroscopes. The norm is often used when the orientation of the sensor is not known. This is common when using a mobile phone or other consumer appliance as a sensor.KeywordsAngular VelocityRecognition RateActivity RecognitionSensor DisplacementAcceleration DataThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

IoT-FAR: A multi-sensor fusion approach for IoT-based firefighting activity recognition

Improving worker safety and productivity is of paramount importance in the manufacturing industry, driving the adoption of advanced sensing and control systems. This concern is particularly relevant within the framework of Industry 5.0. In this context, wearable sensors offer a promising solution by enabling continuous and unobtrusive monitoring of workers’ activities in the manufacturing line. This book chapter focuses on wearable sensor-based human activity recognition and its role in promoting worker safety in manufacturing environments. Specifically, we present a case study on wearable sensor-based worker activity recognition in a manufacturing line with a mobile robot. As wearable sensors comprise various sensor types, we investigate and compare sensor data fusion approaches using neural network models to effectively handle the multimodal sensor data. In addition, we introduce several deep learning-based techniques to improve the performance of human activity recognition. By harnessing wearable sensors for human activity recognition, this book chapter provides valuable insights into improving worker safety on the manufacturing line, aligning with the principles of the Industry 5.0 paradigm. The chapter sheds light on the potential of wearable sensor technologies and offers avenues for future research in this field.

Wearable sensor nodes use activity recognition systems to identify human activities. However, several applications such as physical rehabilitation and professional sports coaching require not only the identification of motion but also quality assessment to provide appropriate feedback to the user. I

Healthcare using body sensor data has been getting huge research attentions by a wide range of researchers because of its good practical applications such as smart health care systems. For instance, smart wearable sensor-based behavior recognition system can observe elderly people in a smart eldercare environment to improve their lifestyle and can also help them by warning about forthcoming unprecedented events such as falls or other health risk, to prolong their independent life. Although there are many ways of using distinguished sensors to observe behavior of people, wearable sensors mostly provide reliable data in this regard to monitor the individual’s functionality and lifestyle. In this paper, we propose a body sensor-based activity modeling and recognition system using time-sequential information-based deep Neural Structured Learning (NSL), a promising deep learning algorithm. First, we obtain data from multiple wearable sensors while the subjects conduct several daily activities. Once the data is collected, the time-sequential information then go through some statistical feature processing. Furthermore, kernel-based discriminant analysis (KDA) is applied to see the better clustering of the features from different activity classes by minimizing inner-class scatterings while maximizing inter-class scatterings of the samples. The robust time-sequential features are then applied with Neural Structured Learning (NSL) based on Long Short-Term Memory (LSTM), for activity modeling. The proposed approach achieved around 99% recall rate on a public dataset. It is also compared to existing different conventional machine learning methods such as typical Deep Belief Network (DBN), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN) where they yielded the maximum recall rate of 94%. Furthermore, a fast and efficient explainable Artificial Intelligence (XAI) algorithm, Local Interpretable Model-Agnostic Explanations (LIME) is used to explain and check the machine learning decisions. The robust activity recognition system can be adopted for understanding peoples' behavior in their daily life in different environments such as homes, clinics, and offices.

Activity recognition (AR) systems for older people are common in residential health care settings such as hospitals and nursing homes, thus numerous methodologies and studies have been developed to improve the effectiveness of AR systems. However, developing sufficiently robust AR systems using sensor data obtained is a challenging task. In this paper, a novel Smart Belt (S-Belt) for Monitoring of elder people using IoT is a smartphone application that predicts a health monitor for elder people and then sends that status together with the health state and anticipated behaviour to the family. Initially, the sensors in the smart belt such as temperature sensor, heartbeat sensor, oxygen sensor, and glucometer are used to gather data from the elders. The details from the sensors will be preprocessed using Stationary wavelet Transformer (SWT) technique and then classified using Multihead-CNN. Then the classified result will be sent to the end users such as doctors and their relatives through a mobile application. The simulation finding shows proposed method performance was evaluated in terms of ACU, PRE, REC, and F1S. The S-Belt model achieves the overall accuracy of 99.64%. MHCNN network improves the accuracy range by 9.5%, 4.42%, and 0.8% better than DNN, RNN, and CNN respectively.

The synergy of communication, computation and sensing capabilities in mobile systems-on-chip artifacts such as smartphones has made possible the development of wearable smart sensor systems for user activity monitoring and recognition. Assessing physical activity is useful to enhance people health as experts have evidenced a clear correlation between physical activity and overweight, obesity and metabolism-related syndromes. Due to user acceptability and the convenient nonintrusive manner for measuring data, smartphones have the advantage of taking proprioceptive motion measurements outside of a controlled environment for rather long periods of time using embedded sensors such as the accelerometer, however, activity recognition poses several challenges. Particularly, though work has been reported for accelerometer-based activity recognition using smartphones, the portability of the device to a single fixed tight position has been a major constraint to easy the interpretation of the collected data on resource-limited devices. In this chapter, a human activity hierarchical recognition system based on neural networks without the need of the smartphone to be constrained to a single fixed position is presented. Yet it is used as a representative example to show the challenges, the role and some of the main potential impacts that smartphones have and will have in mHealth. Experimental results on Android-capable smartphones on four on-body locations show that the recognition system achieves high classification rates, above 92%, for five activities including static, walking, running, and up-down stairs walking, which outperforms other proposals. The system is fully implemented in a smartphone running continuously in near real-time with reduced power consumption in a proof-of-concept client-server application for mHealth.

Human activity recognition is one of the most challenging research areas in human-machine interaction. Activity recognition by wearable sensors produces significant information about a person's functional ability and lifestyle. In this paper, we present an inertial sensors-based approach for activity and postural transition recognition. Efficient features are extracted from the sensor data. Then, the features are normalized using Z-score followed by Generalized Discriminant Analysis (GDA) to make them more robust. Finally, the features are trained with a Recurrent Neural Network (RNN) for activity and postural transition recognition. The proposed method recognizes 12 activities with an overall accuracy of 97%. In the activity recognition system, the multimodal input data is collected from only one accelerometer and one gyroscope sensor embedded in the user's smartphone. The proposed system can contribute in research fields such as human-robot interaction for practical applications such as eldercare in smart homes.

PurposeThe emergence of smart wearables using clothing as a technology platform is a significant milestone with considerable implications for industrial convergence, creating new value for fashion. This paper aimed to present a premeditated prototype to integrate a human activity recognition (HAR) system into outdoor clothing.Design/methodology/approachFor the development of wearable HAR (WHAR) clothing, this paper explored three subject areas: fashion design related to the structural feature of the clothing platform, electronics related to wearable circuits and modules design and graphic user interface design related to smartphone application development.FindingsFor WHAR functions in outdoor terrains, the coexistence of accelerometer–gyroscope sensing and distance-sensing could be practical to surpass the technological limitation of activity and posture recognition with gyro sensors highly depending on the changes of acceleration and angles.Research limitations/implicationsThrough the vital sign check and physical activity–change recognition function, this study's WHAR system allows users to check their health by themselves and avoid overwork. A quick rescue is possible manually and automatically in a dangerous situation by notifying others. Thus, it can help protect users' health and safety (life).Originality/valueThis study designed the modularization of HAR functions generally installed in indoor medical spaces. Through the approach, smart clothing–embracing WHAR systems optimized for health and safety care for outdoor environments was pursued to diversify expensive roles of clothing for technological applications.

In spite of the importance of activity recognition (AR) for intelligent human-computer interaction in emerging smart space applications, state-of-the-art AR technology is not ready or adequate for real-world deployments due to its insufficient accuracy. The accuracy limitation is directly attributed to uncertainties stemming from multiple sources in the AR system. Hence, one of the major goals of AR research is to improve system accuracy by minimizing or managing the uncertainties encountered throughout the AR process. As we cannot manage uncertainties well without measuring them, we must first quantify their impact. Nevertheless, such a quantification process is very challenging given that uncertainties come from diverse and heterogeneous sources. In this article, we propose an approach, which can account for multiple uncertainty sources and assess their impact on AR systems. We introduce several metrics to quantify the various uncertainties and their impact. We then conduct a quantitative impact analysis of uncertainties utilizing data collected from actual smart spaces that we have instrumented. The analysis is intended to serve as groundwork for developing “diagnostic” accuracy measures of AR systems capable of pinpointing the sources of accuracy loss. This is to be contrasted with the currently used accuracy measures.

Human activity recognition is a key technology for ICT-based (infomation and communication technologies) assistive applications. The most successful activity recognition systems for intelligent environments in terms of performance rely on supervised learning techniques. However, those techniques demand large labelled data sets for specific sensor deployments and monitored person. Such requirements make supervised learning techniques not to scale well to real world deployments, where different sensor infrastructures may be used to monitor different users. In this paper, we present a novel activity recognition system, based on a combination of unsupervised learning techniques and knowledge-based activity models. First, we use a domain-specific data mining algorithm previously developed by Cook et al. to extract the most frequent action sequences executed by a person. Second, we insert knowledge-based activity models in a novel matching algorithm with the aim of inferring what activities are being performed in a given action sequence. The approach results on a scalable activity recognition system, which has been tested on three real data sets. The obtained performance is comparable to supervised learning techniques.

Human activity recognition is a basic building block in numerous healthcare systems, mainly because the ability to understand the user's situation and context. This paper presents a solution to the general problem with evaluation of human activity recognition systems, i.e., an activity recognition system may perform perfectly in controlled lab experiments, but significantly worse once applied to more realistic conditions. The solution is presented through the practical experience gained with the creation of our RAReFall activity recognition system. Although the system was awarded the first place at the EvAAL-AR live competition, the recognition accuracy at the competition was significantly lower compared to the controlled lab experiments performed just before the competition. To overcome the encountered problem we developed an automatic calibration method, which solves the encountered problem by adapting and re-calibrating the accelerometer data in real-time while the user is performing everyday activities. The method increased the overall accuracy for 8 percentage points and for 51 percentage points for the sitting activity.

Comprehensive and quantitative assessment of human physical activity in daily life is valuable for healthcare, especially for those who suffer from obesity and neurological disorders or are at high risk of dementia. Common wearable devices, e.g., smartwatches, are insufficient and inaccurate for monitoring highly dynamic limb movements and assessing human motion. Here, we report a new wearable leg movement monitoring system incorporating a custom-made motion sensor with machine learning algorithm to perceive human motion accurately and comprehensively during diverse walking and running actions. The system enables real-time multimodal perceptions of personal identity, motion state, locomotion speed, and energy expenditure for wearers. A general law of extracting real-time metabolic energy from leg movements is verified although individual gaits show differences. In addition, we propose a novel sensing configuration combining unilateral lower leg movement velocity with its angular rate to achieve high accuracy and good generalizability while simplifying the wearable system. Advanced performances in personal identification (accuracy of 98.7%) and motion-state recognition (accuracy of 93.7%) are demonstrated. The wearable system also exhibites high-precision real-time estimations of locomotion speed (error of 3.04% to 9.68%) and metabolic energy (error of 4.18% to 14.71%) for new subjects across various time-varying conditions. The wearable system allows reliable leg movement monitoring and quantitative assessment of bodily kinematic and kinetic behaviors during daily activities, as well as safe identity authentication by gait parameters, which would greatly facilitate smart life, personal healthcare, and rehabilitation training.

This Letter investigates and reports on a number of activity recognition methods for a wearable sensor system. The authors apply three methods for data transmission, namely 'stream-based', 'feature-based' and 'threshold-based' scenarios to study the accuracy against energy efficiency of transmission and processing power that affects the mote's battery lifetime. They also report on the impact of variation of sampling frequency and data transmission rate on energy consumption of motes for each method. This study leads us to propose a cross-layer optimisation of an activity recognition system for provisioning acceptable levels of accuracy and energy efficiency.