Drift-Free Inertial Sensor-Based Joint Kinematics for Long-Term Arbitrary Movements

Abstract

The ability to capture joint kinematics in outside-laboratory environments is clinically relevant. In order to estimate kinematics, inertial measurement units can be attached to body segments and their absolute orientations can be estimated. However, the heading part of such orientation estimates is known to drift over time, resulting in drifting joint kinematics. This study proposes a novel joint kinematic estimation method that tightly incorporates the connection between adjacent segments within a sensor fusion algorithm, to obtain drift-free joint kinematics. Drift in the joint kinematics is eliminated solely by utilizing common information in the accelerometer and gyroscope measurements of sensors placed on connecting segments. Both an optimization-based smoothing and a filtering approach were implemented. Validity was assessed on a robotic manipulator under varying measurement durations and movement excitations. Standard deviations of the estimated relative sensor orientations were below 0.89 • in an optimization-based smoothing implementation for all robot trials. The filtering implementation yielded similar results after convergence. The method is proven to be applicable in biomechanics, with a prolonged gait trial of 7 minutes on 11 healthy subjects. Three-dimensional knee joint angles were estimated, with mean RMS errors of 2.14 • , 1.85 • , 3.66 • in an optimization-based smoothing implementation and mean RMS errors of 3.08 • , 2.42 • , 4.47 • in a filtering implementation, with respect to a golden standard optical motion capture reference system. Tommy Verbeerst received the M.Sc. degree in electrical engineering from KHBO, Ostend, Belgium, in 2008. Since 2013, he has been working with KU Leuven and UC Vives. He is affiliated with the Department of Electrical Engineering (ESAT), KU Leuven Campus Bruges, Belgium. His current research interest includes the fields of engineering education, robotics, and machine-vision. Mark Versteyhe received the M.Sc. degree in mechanical engineering and the Ph.D. degree in applied sciences, from KU Leuven in 1995 and 2000, respetively.He has worked 16 years in industry in various functions linked to research and innovation. Since October 2016, he has been a Professor with KU Leuven's Faculty of Engineering Technology, Technology Campus Brugge, where he co-ordinates the research effort on connected mechatronics. His research focus lies in studying and applying the holistic paradigm of mechatronic system design. His special interest goes to "Dependability" which encompasses reliabilityavailability-robustness and security of a system and "Distributed Systems" which are treated as a complex ecosystem of machines and humans that are connected within the Industry 4.0 paradigm shift.Kurt Claeys received the M.Sc. degree in musculoskeletal rehabilitation sciences and physiotherapy from the University of Ghent, Belgium, in 1993, and the Ph.D. degree in orthopedic manual therapist from the IRSK-WINGS institute Ieper, Belgium, in 2005, and the Ph.D. degree from KU Leuven, Belgium,...

OpenSense: An open-source toolbox for Inertial-Measurement-Unit-based measurement of lower extremity kinematics over long durations

“…These implicit corrections by the sensor fusion algorithm occurred at different times depending on the activity, but for each of the movement trials (continuous walking and a sequence of movements), the frequency and the duration of periods of low angular velocity were sufficient to mitigate drift. The approach further has the benefit of being activity-agnostic, compared to some previous approaches that were tailored to specific activities (e.g., (36) for running; (37) for specific phases during walking) or reliant on achieving relatively high joint center accelerations (23,38). We also verified that by selecting parameters for the sensor fusion algorithm that mitigated drift, we did not sacrifice accuracy over short durations.…”

Magnetometer Robust Deep Human Pose Regression With Uncertainty Prediction Using Sparse Body Worn Magnetic Inertial Measurement Units

“…However when MEMS accelerometers are integrated with rate gyros and magnetometers in a magnetic-inertial measurement units (MIMU) or 9-axis IMU, sensor fusion [28] can be used to obtain full 3D orientation (pitch, roll and heading). More recently 9-axis or 6-axis IMUs (less magnetometers) have been used [2,3,5,23,24] for full body human motion capture. The human body is assumed as comprising of rigid segments articulated at joints and one sensor per segment is sufficient to compute 3D joint angle if adjacent segment orientations as rigid body are known.…”

“…The human body has constrained degree of freedom and temporal coherence and smoothness is an important feature of human motion. Many existing kinematic or inverse kinematic based i-Mocap frameworks, therefore uses predefined constraints to reduce measurement errors or drifts [2][3][4][5][6]. In past research [7][8][9], a small set of inertial sensors is shown to estimate 3D pose to a reasonable accuracy.…”

“…To the best of our knowledge, our work demonstrates a magnetometer robust 3D human pose estimation using reduced or sparse sensors for the first time. Previous work [2,3,5,23,24] has addressed this problem in the context of full body worn sensors (typical 13-17 sensors).…”

Machine Learning to Improve Orientation Estimation in Sports Situations Challenging for Inertial Sensor Use

“…These sources provide additional (indirect) information about the sensor orientation. Another (magnetometer-free) solution that is used to detect the direction of gravity without assuming that only gravity is measured is to attach multiple sensors on connected body segments (Lee and Jeon, 2019;Weygers et al, 2020). Although those approaches previously produced accurate orientation estimates (Brodie et al, 2008;Zhang et al, 2016), the benefits of using a single sensor (easy to use and cheap) diminish.…”