Biomechanical energy harvesting converting kinetic energy from human motion into electrical energy appeared as a promising technology for powering mobile devices. The optimization of these energy harvesting systems requires knowledge about human gait as an energy source to maximize the power output. Therefore, motions during walking and running have to be evaluated to determine the amount of available and convertible energy in several body parts. In case of vibration energy harvesting systems, their efficiency is also dependent on the adjustment of the systems' frequency and dimensions and frequency characteristics to the driving force. Thus, this paper presents a solution fusing optical sensors and inertial measurement units (IMU) to analyze human locomotion. High-speed cameras are used to track the positions and angles of the synchronized inertial body sensors in an earth frame while the IMUs acquire gyroscope, accelerometer and magnetometer data. This data is used to calculate linear acceleration and actual orientation represented in quaternions applying an algorithm by Madgwick. The orientation gives major information about the effect of the driving force on moving masses of the system as energy harvesting devices are often designed as single-axis generators. Furthermore, frequency responses of acceleration data from different body positions while gait cycles are analyzed. Finally, prospects and issues converting acceleration into velocity and position data and vice versa are discussed.
|