Joint Angle Estimation with Customized Wristband Based on Online Incremental Learning

2026-05-28

AI Summary

This work addresses the challenge of adapting traditional wearable sensors to inter-user variability and inconsistent wear conditions in wrist joint angle estimation. To overcome this, the authors propose a two-stage online incremental learning framework. In the first stage, inertial measurement unit (IMU) data serve as self-supervised ground truth to continuously update the model according to the user’s motion characteristics. The second stage leverages a custom-designed flexible wristband for real-time angle estimation, enabling on-the-fly training and immediate deployment—effectively realizing “train-as-you-collect, use-as-you-train.” This approach robustly mitigates data drift caused by variations in sensor placement, handedness, and individual biomechanics, substantially enhancing generalization and practicality. Experimental results demonstrate an average joint trajectory estimation error of approximately 15° across diverse scenarios, confirming the system’s robustness under varying strain conditions.

Abstract

Intelligent wearable technology plays an increasingly important role in human-computer interaction, motion, and health monitoring. To ensure comfort and practicality of use, one common form for motion monitoring is to utilize soft wearable sensors. However, many research applications regarding wearable sensors are simplistic and difficult to adapt to different situations. This study proposes a system for estimating the angle of the wrist joint using a customized wristband based on an online incremental learning approach. It is a two-stage estimation method: the first stage updates the model based on the wearer's wrist movement characteristics using online learning, integrating real-time data from an IMU as ground truth. The second stage utilizes the updated model for estimation of wrist joint angle solely with the wristband. In other words, model training is completed during data acquisition, allowing the trained model to be used for subsequent angle estimation. This method offers advantages in adapting to data drift caused by variations in different testing configurations, such as the left and right wrists of the same subject, deviations in the wearing position on the same wrist, and even differences among various subjects. The results indicate that the sensors exhibit good performance under strain variations, and the wrist joint trajectory estimation of the proposed system has an approximate error of 15 degree in different scenarios.

Problem

• joint angle estimation
• wearable sensors
• data drift
• human-computer interaction
• motion monitoring

Innovation

• online incremental learning
• wearable sensors
• joint angle estimation
• data drift adaptation
• IMU-based ground truth