Wearable technology continues to evolve, offering innovative ways to monitor fitness and performance. One exciting development is the measurement of kinetic power, commonly used in cycling but less explored in running and walking.
In his research, Lambert Brink under supervision of Dr Willem Smit, explores the creation of an embedded foot device designed to measure kinetic power during these activities—aiming to deliver more accurate and consistent data than existing marketplace models.
The Problem with Current Power Measurement Devices
Despite the availability of devices like the Stryd pod and Garmin’s lineup, measuring kinetic power during walking remains inconsistent. These gadgets often provide conflicting results, lacking the reliability seen in cycling power meters. Currently, no standardized method exists to accurately measure power during walking and running, leaving users with incomplete or inconsistent data.
This research tackles the challenge head-on by developing a low-power foot device that gathers crucial biomechanical data like gait, velocity, and position. The goal? To create a dependable system that delivers consistent, reliable kinetic power measurements across a range of walking and running activities.
Cutting-Edge Methodology: A Fusion of Hardware and Algorithms
The heart of this research is a custom-designed foot device powered by an Inertial Measurement Unit (IMU) and a Bluetooth Low Energy (BLE) transmitter. These components capture essential data on acceleration, angular velocity, and magnetic flux, which are then processed using a quaternion-based algorithm to calculate orientation. The algorithm’s accuracy ensures precise velocity measurements, aided by a Zero Velocity Update (ZVU) process that eliminates drift.
The research incorporates several filtering methods to tackle the inevitable noise in acceleration data, ensuring clean, accurate readings. The data is then stored on flash memory or transmitted wirelessly for real-time analysis, making the system both flexible and powerful.
During testing, the foot device demonstrated excellent accuracy in orientation and velocity calculations. Although there were some hiccups in data storage and step detection, the overall design showed promise for providing reliable kinetic power measurements through everyday biomechanical movements.
Results That Move the Needle
The project delivered a working prototype that excelled in measuring kinetic power during walking. Orientation calculations were spot-on, and velocity and position estimates were highly accurate—within 10% of testing standards. Despite occasional challenges with step detection, the device proved it could deliver valuable insights into biomechanical performance.
One surprising outcome was the inconsistency in the Stryd pod’s results when used for walking, which confirmed that current marketplace models still fall short in accuracy. This reinforces the need for ongoing refinement and development in kinetic power measurement technology.
Figure 1: Embedded design of the system
What’s Next? The Future of Kinetic Power Measurement
Looking ahead, there are exciting opportunities to build on this research. The next steps involve improving memory management to enhance step detection and reduce errors during data storage. Incorporating multi-threaded processing will enable the device to handle data storage and measurement simultaneously, eliminating some of the issues encountered in testing.
Further, the kinetic power algorithm can be expanded to consider environmental factors like terrain gradient and wind speed, paving the way for more comprehensive outdoor tests. Including running scenarios in future studies will also broaden the device’s potential applications, making it a versatile tool for athletes and casual users alike.
With continued development, this research could help establish an open-source standard for kinetic power measurement, paving the way for more accurate, reliable fitness devices that meet the demands of modern users.
Download and read the complete research at https://scholar.sun.ac.za/items/feb13842-e32a-4322-843e-9340f5eaaaa0
