Research On Attitude Algorithm Based On Micro Inertial Sensors

With the rapid development and wide application of micro inertialsensor, the micro inertial measurement system consisted of the micro inertialsensor becomes a hot research field, and is the inertial measurement systemdevelopment direction which is low cost, precision, small size, low-powerdissipation, high overload and high reliability. Due to low cost, accuracy,small size and etc., the micro inertial attitude measuring system is widelyused in aerospace, automotive, robot and many other fields.This paper has done in-depth research on the sensor error analysis andthe key technology of attitude algorithm in the micro inertial technology field.By using the Micro Inertial Measurement Unit of VN-100, it also has doneexperiments and simulation. The main research is as follows:Firstly, it the development of micro inertial sensor and micro inertialmeasurement system is introduced. The principle of strapdown inertialnavigation and traditional attitude algorithms are elaborated in detail.Secondly, the principles and technical parameters of MEMS gyros, MEMS accelerometers and magnetic sensor are introduced. And the sensorerror models are analyzed and build. Then the gyro random drift error isanalyzed by Allan variance method.Finally, the Complementary Filter and typical Kalman Filter aredesigned separately and used to filtering the outputs of accelerometer andmagnetic sensor. This improves the accuracy of attitude angles calculated bythe algorithm based on the outputs of accelerometer and magnetic sensor. Inorder to get attitude quarternion error estimates and compensate for attitudequarternion updated immediately, the Square-root Cubature Kalman Filter isdesigned using the outputs of accelerometer and magnetic sensor tocompensate the gyro bias.