Simulation of inertial sensors in virtual environments

This study deals with the simulation of inertial and magnetic sensors in virtual environment systems. We proposed a mathematical model for the correction of accelerometer, gyroscope, and magnetometer readings to account for the effects of temperature, cross-coupling, and noise. We also proposed using the sensor readings to compute the Euler angles that define the vehicle's orientation. We calibrated the accelerometer and gyroscope readings to reduce the zero drift error. The Euler angles were derived by integrating the gyroscope readings and using the found relationships between the accelerometer and magnetometer readings. To eliminate the cumulative integration error caused by sensor noise, we implemented a complementary filter. It mixes the angles in a given ratio. We tested these approaches in a virtual environment with a virtual quadcopter and a coaxial-rotor helicopter. The results were promising and can be used in advanced simulators designed to train UAV operators.