Sensors-based data fusion solution design for 3D motion estimation with application in bio-logging

The main purpose of this paper deals with the orientation estimation problem of a rigid-body motion in space. We present an algorithm for attitude estimation, expressed in quaternion representation, using low-cost sensors as 3-axis accelerometer, 3-axis magnetometer and 3-axis gyroscope. The algorithm is based on a complementary nonlinear observer coupled with a Levenberg Marquardt Algorithm (LMA). Moreover, the proposed solution exploits kinematic equation model and includes the estimation of rate gyros biases to compensate angular velocity measurements. This algorithm is developed in order to address the well-known problem of the weak dynamics of the attitude sensors (accelerometer and magnetometer). The efficiency of the proposed observer is illustrated by an attitude estimation example in presence of realistic measurements provided by low-cost sensors. Some preliminary experimental results are provided also to prove the performance of the proposed method. The developed approach will be applied in future works in Bio-logging area which interests to study the animal behavior and its energy expenditure by determining its movement patterns (3D motion or orientation).