Uploaded May 2018 | Updated September 2026, 2 weeks ago
ICRA 2018 Spotlight Video
Interactive Session Thu PM Pod C.1
Authors: Xiao, Xuesu; Zarar, Shuayb
Title: Machine Learning for Placement-Insensitive Inertial Motion Capture
Abstract:
Although existing inertial motion-capture systems work reasonably well (less than 10 degrees error in Euler angles), their accuracy suffers when sensor positions change relative to the associated body segments (positive minus 60 degrees mean error and 120 degrees standard deviation). We attribute this performance degradation to undermined calibration values, sensor movement latency and displacement offsets. The latter specifically leads to incongruent rotation matrices in kinematic algorithms that rely on rotational transformations. To overcome these limitations, we propose to employ machine-learning techniques. In particular, we use multi-layer perceptrons to learn sensor-displacement patterns based on 3 hours of motion data collected from 12 test subjects in the lab over 215 trials. Furthermore, to compensate for calibration and latency errors, we directly process sensor data with deep neural networks and estimate the joint angles. Based on these approaches, we demonstrate up to 69% reduction in tracking errors.
ICRA 2018 Spotlight Video
Interactive Session Thu PM Pod C.1
Authors: Xiao, Xuesu; Zarar, Shuayb
Title: Machine Learning for Placement-Insensitive Inertial Motion Capture
Abstract:
Although existing inertial motion-capture systems work reasonably well (less than 10 degrees error in Euler angles), their accuracy suffers when sensor positions change relative to the associated body segments (positive minus 60 degrees mean error and 120 degrees standard deviation). We attribute this performance degradation to undermined calibration values, sensor movement latency and displacement offsets. The latter specifically leads to incongruent rotation matrices in kinematic algorithms that rely on rotational transformations. To overcome these limitations, we propose to employ machine-learning techniques. In particular, we use multi-layer perceptrons to learn sensor-displacement patterns based on 3 hours of motion data collected from 12 test subjects in the lab over 215 trials. Furthermore, to compensate for calibration and latency errors, we directly process sensor data with deep neural networks and estimate the joint angles. Based on these approaches, we demonstrate up to 69% reduction in tracking errors.
