Motion tracking algorithms for inertial measurement

Abstract

In this paper, we describe the development of the software algorithms required to interpret sensor data developed by a wearable miniaturized wireless inertial measurement unit (IMU) to enable tracking of movement Traditionally, inertial tracking has involved the use of off the shelf motion sensors in the form of an inertial measurement unit, in combination with a GPS based receiver system for improved accuracy. Several immediate concerns are evident when a low cost, low power consumption, miniaturised solution is needed in applications such as animal tracking. GPS solutions have proven to be costly requiring an expensive satellite link & entail power supply and size concerns when deployed on live animals. In particular applications, GPS coverage is not available for all application scenarios and alternative mechanisms for motion tracking are required. IMUs cannot be used in isolation for absolute position tracking, since an IMU calculates position utilising a square function of time (t) where the error is proportional to the sampling time, any errors in the output of the sensors are therefore also multiplied by t². This typically leads to large positional errors in operation: These issues can potentially be addressed by using only a low cost modular IMU solution to enable the mapping of movement if appropriate algorithms are implemented The goal of this paper is to present a mathematical algorithm that enables an inertial-based tracking system to be realized. This algorithm could then be used with GPS (GPS is commonly used but there are other methods like triangulation) along with a Kalman Filter algorithm providing an accurate 3-Dimensional tracking system.

Unobstructive Body Area Networks (BAN) for Efficient Movement Monitoring

“…For these tests were implemented simple versions of the Kalman filter (KF), as well as, more advanced versions of the extended Kalman filter (EKF) specially designed for fusing data from inertial sensors [29,30]. …”

“…The third phase uses the same number of sensor nodes, but this time, before the data is applied to the CGHD, it is filtered in order to remove as much noise as possible. For these tests were implemented simple versions of the Kalman filter (KF), as well as, more advanced versions of the extended Kalman filter (EKF) specially designed for fusing data from inertial sensors [ 29 , 30 ].…”

Self-Driving Algorithm and Location Estimation Method for Small Environmental Monitoring Robot in Underground Mines

“…g represents gravity acceleration, while a Ugravity , a Vgravity , a Wgravity represents gravity acceleration vectors in the U, V, and W axes. a Ucentripetal , a Vcentripetal , a Wcentripetal denotes centripetal in the direction of U, V, and W, w U , w V , w W , and V U , V V , V W , denote the angular and linear velocities in the U, V and W axes [50].…”

A Manual Actions Expressive System (MAES)

“…Pointing-At (Torre, Torres and Fernstrom 2008; Torre 2013) measures 3D orientation ( attitude ) with high accuracy. A bending sensor on the index finger behaves as a three-state switch (as opposed to MAES's continuous bending data in all five fingers).…”