MEMS Gyro Noise Estimation and Modeling for Precise Navigation Simulation

Inertial navigation system (INS) pivotal error sources are related to the deterministic errors and random errors of the inertial measurement unit (IMU). Low-cost inertial sensors are distinguished by high noise and significant performance uncertainties. Consequently, errors related to low-cost INS states (position, velocity, and attitudes) are rising rapidly in stand-alone mode. If accurate performance can be achieved with low-cost IMU, the cost of real applications can be reduced and the development of new applications may be made feasible. Errors in gyroscopes (gyro) which are the main building block of IMU play a significant role, while the errors in the accelerometers can be calibrated and compensated with reasonable precision by comparison with the gravity of the earth. So precise modeling of gyro for navigation simulation for missiles and unmanned vehicles is essential. To achieve this goal, firstly an error estimation algorithm is implemented. This algorithm firstly is tested with a simulated signal with a known noise parameter injected in it. Moreover, the calibration of deterministic errors is performed. Furthermore, a study of the autocorrelation function of the sensor outputs is presented, and a method for calculating the frequency response to determine potential stability problems is provided by calculating the cutoff frequency. The typical analysis methods discussed throughout the paper are intended to assist designers to develop or choose gyro that is uniquely adapted to the design application requirements. The development technique is used to characterize a tactical grade Micro-Electro-Mechanical Systems (MEMSIC) IMU, and the test results prove its accuracy comparing to the manufacturing calibration sheet.