Feasibility of Tracking Human Kinematics with Simultaneous Localization and Mapping (SLAM)

Area of Science:

• Biomedical Engineering
• Robotics
• Computer Vision

Background:

• Human kinematics tracking is crucial for various applications, including rehabilitation, sports science, and robotics.
• Existing technologies for human motion tracking have limitations in accuracy, cost, or setup complexity.

Purpose of the Study:

• To evaluate a novel wearable technology integrating inertial sensors and cameras for human kinematics tracking.
• To assess the accuracy and practical applicability of on-board Simultaneous Localization and Mapping (SLAM) algorithms in this wearable system.

Main Methods:

• Development and testing of a wearable device combining inertial measurement units (IMUs) and cameras.
• Implementation of on-board SLAM algorithms for real-time camera localization within the environment.
• Quantitative comparison of the system's kinematic tracking data against a robotic arm as a ground truth.

Main Results:

• The wearable system demonstrated high accuracy in tracking human kinematics.
• Orientation error was consistently less than 1 degree.
• Position error was less than 4 centimeters when compared to the robotic arm's measurements.

Conclusions:

• The integrated wearable technology effectively tracks human kinematics with high precision.
• The accuracy achieved by the SLAM algorithms is sufficient for numerous practical applications.
• This technology offers a promising alternative to existing methods, potentially overcoming their limitations.