MEMS-based low-cost strap-down AHRS research

With the rapid development of modern Micro-Electro-Mechanical Systems (MEMS) technology, the requirement of low cost strapdown AHRS based on MEMS sensor is becoming feasible. This paper presents the results of the attitude determination using low-cost MEMS-based sensor. Aiming at the accumulative deviation incurring by low-cost MEMS gyroscope drifting and integration deviation, this study proposes a multi-sensor data fusion algorithm that uses benchmark reference vectors to detect and correct the gyro drift. The proposed algorithm adopts the method of attitude evaluation based on Direction Cosine Matrix (DCM). A BP neural network PID controller is used to compute gyro drift for real-time correction. Numerical simulation shows that the algorithm meets the requirements of the system. In order to test the actual performance of the system, a hardware platform of strap-down AHRS is designed, based on a three-axis gyroscope, a three-axis accelerometer, three-axis magnetic sensors and a Digital Signal Processor (DSP). Finally, the three-axis flight turntable table test results obtained, show that the static accuracy of the attitude angle is better than 0.5 degrees and the dynamic accuracy is better than 3 degrees. Moreover, the AHRS is tested on a micro-Unmanned Aerial Vehicle (UAV), and the practical application feasibility is validated. (C) 2014 Elsevier Ltd. All rights reserved.