Details of a Proven Software for the Control of Unmanned Aircraft Systems - Part 1 - Determining the Orientation

The paper presents software components that have been successfully used over many years to control unmanned aircraft systems, with a focus in this first part on attitude (orientation) determination. It explains the mathematical foundations of orientation representation, including quaternions, rotation matrices, and a small-angle simplification of the rotation matrix suitable for high-frequency flight control loops. The work addresses the limitations and error characteristics of accelerometers and gyroscopes and motivates combining their data rather than relying on single sensors. Two proven orientation estimation methods are detailed: a complementary filter that pairs a low-pass filter on accelerometer data with a high-pass filter on gyro data, and a data fusion approach that blends a gyro-based orientation estimate with an accelerometer-based measurement. The described algorithms are designed to be computationally lightweight so they can run on simple processors while still providing robust and stable orientation information for multicopter flight controllers.