A Miniature Integrated Navigation System for Rotary-Wing Unmanned Aerial Vehicles

Abstract EN

This paper presents the development of a lowcost miniature navigation system for autonomous flying rotary-wingunmanned aerial vehicles (UAVs).

The system incorporatesmeasurements from a low cost single point GPS and a triaxialsolid state inertial/magnetic sensor unit.

The navigation algorithmis composed of three modules running on a microcontroller:the sensor calibration module, the attitude estimator, and thevelocity and position estimator.

The sensor calibration modulerelies on a recursive least square based ellipsoid hypothesiscalibration algorithm to estimate biases and scale factors ofaccelerometers and magnetometers without any additional calibrationequipment.

The attitude estimator is a low computationallinear attitude fusion algorithm that effectively incorporates highfrequency components of gyros and low frequency components ofaccelerometers and magnetometers to guarantee both accuracyand bandwidth of attitude estimation.

The velocity and positionestimator uses two cascaded complementary filters which fusetranslational acceleration, GPS velocity, and position to improvethe bandwidth of velocity and position.

The designed navigationsystem is feasible for miniature UAVs due to its low cost, simplicity,miniaturization, and guaranteed estimation errors.

Bothground tests and autonomous flight tests of miniature unmannedhelicopter and quadrotor have shown the effectiveness of theproposed system, demonstrating its promise in UAV systems.