Real Time Estimation of rigid body orientation based on inertial sensors measurements

This paper presents a programmable DSP board developed for real-time estimation of rigid body orientation. This work takes place in a research project carried out by students. The project consists of unmanned aerial vehicle’s (UAV) attitude estimation. This estimation is processed by an Inertial Reference System (IRS) including the TMS320F2812 processor, low cost integrated Micro Electro Mechanical Systems (iMEMS) such as accelerometers and gyroscopes, temperature sensors, and magnetometers. Gyroscopes provide angular velocities that can be integrated to yield orientation. Accelerometers and magnetometers, pointing to gravity direction and north direction enables drift compensation. So the system performs a drift-free attitude estimation using quaternions for rotations representation. One benefit of this system is that it enables high integration by exploiting the processor’s capacities for sensors interfacing. Moreover it takes advantage of the high performance processor to achieve a real time estimation of the body’s orientation. For that digital processing includes signals filtering, measurements compensation and a lot of trigonometric functions for Euler angles computing. The system performances can be measured using a PC for results visualisation. The system has been designed to integrate in the future GPS component and pressure sensor. These components will allow more accurate attitude estimation and positioning.