An Ensemble Bayesian Dynamic Linear Model for Human Activity Recognition

• Published in: IEEE access Vol. 13; p. 1
• Main Authors: Pitombeira-Neto, Anselmo R., De Franca, Diego S., Cruz, Livia A., Da Silva, Ticiana L. C., De Macedo, Jose Antonio F.
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.01.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerometers; Bayes methods; Bayesian analysis; Bayesian dynamic linear models; Classification; Computational modeling; Data acquisition; Data models; Feature extraction; Hidden Markov models; Human activity recognition; inertial measurement unit; Inertial platforms; Kalman filter; Nusselt number; Sensor phenomena and characterization; Sensors; Time series; Time series analysis
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2025.3541385

Human activity recognition (HAR) has been gaining attention in recent years as a result of its many applications in health, sports, entertainment, and surveillance. Due to the current ubiquity of inertial measurement units (IMUs) in consumer electronics, including accelerometers and gyroscopes, HAR applications have increasingly used signals produced by these sensors. However, HAR from IMU data is challenging, since time-series data generated from human activity are typically multivariate, non-stationary, and noisy. In this work, we investigate the application of Bayesian dynamic linear models (BDLMs) to the online classification of time-series data of human activity acquired from IMUs. BDLMs are promising in HAR from IMU signals because they seamlessly handle temporal dependencies and uncertainty inherent in sensor data. Unlike static classifiers, BDLMs account for the sequential nature of IMU signals, enabling more accurate tracking of transitions between activities. In particular, we propose a method based on an ensemble BDLM for online HAR that is fully transparent and requires little preprocessing of data. We test the proposed method in two tasks, activity classification and intensity classification, and use two real datasets with diverse activities and subjects. The experimental results indicate that the proposed ensemble BDLM is competitive with consolidated benchmark methods and can be an effective method in real applications of online HAR.