Sensor data fusion for attitude stabilization in a low cost Quadrotor system

With the development of low-cost and lightweight Micro-Electro-Mechanical System (MEMS) sensors, it has become possible to efficiently develop an attitude estimator, thus renewing the interests into vertical-take-off and landing (VTOL) aircrafts. A quadrotor with an Attitude Heading Reference System (AHRS) was built from scratch for investigation of heterogeneous sensor data fusion and attitude stabilization. This paper focuses on the system integration through the use of a 16-bit embedded microcontroller and sensor data fusion approach to reduce the gyroscope drift and the impact of vibration noise on measurements. The Quadrotor was able to hover over the ground stably indoors and react to applied disturbances. Attitude stabilization with the implementation of a Kalman Filter and PID controller is verified to be satisfactory through tests and response observed in actual flight data.