UAV Sensors Autonomous Integrity Monitoring—SAIM

Abstract

For Unmanned Aerial Vehicles (UAVs), it is of crucial importance to develop a technically advanced Collision Avoidance System (CAS). Such a system must necessarily consist of many sensors of various types, each one having special characteristics and performance. The poor performance of one of the sensors can lead to a total failure in collision avoidance if there is no provision for the performance of each separate sensor to be continuously monitored. In this work, a Sensor Autonomous Integrity Monitoring (SAIM) methodology is proposed. The configuration of the sensors and their interaction is based on a fusion procedure that involves a total of five sensors. Accordingly, the performance of each one of the sensors is continuously checked against the combined (fused) operation of the other four. A complementary experiment with a total of four sensors, one of which had low performance, was also conducted. Experimental results reveal a reliable approach for Sensor Autonomous Integrity Monitoring (SAIM). The method can be easily extended to a larger number of sensors.