Design and Validation of an IMU Based Full Hand Kinematic Measurement System

Abstract

Measurement of hand kinematics is commonly done using data gloves and optical trackers and finds application in biomechanics, motor control research, clinical assessment, virtual reality etc. While data gloves are expensive and restrict the dexterity of the hand, optical trackers are susceptible to the line-of-sight problem and can be used only in a laboratory setting. Over the last decade, the use of Inertial Measurement Units (IMUs) to measure kinematics has gained traction due to their affordability and good accuracy. This paper presents the design and validation of a BNO055 IMU-based full Hand Kinematics Measurement System (HKMS). The best features from existing IMU-based devices were identified from the literature and incorporated. The HKMS streams orientation information of 16 BNO055 IMUs in real-time at 100 Hz via Wi-Fi or USB. A rigorous static and dynamic validation of the BNO055 IMU was done against 3D printed models and the highly accurate Electro Magnetic Tracking System (EMTS). The RMSE errors were found to be acceptable for the measurement of hand kinematics. Two experiments were conducted to collect hand kinematic data for various postures and object grasps using the HKMS and EMTS, respectively. Synergy analysis was done and range of motion of the joints was calculated using the two datasets. The results were compared to get an idea of the quality of the dataset collected using the HKMS. The results as well as the overall validation of the sensors indicate that the HKMS may be suitable for usage in a laboratory or clinical setup.