Unmanned Systems Technology 022 | XOcean XO-450 l Radar systems l Space vehicles insight l Small Robot l BMPower FCPS l Prismatic HALE UAV l InterDrone 2018 show report l UpVision l Navigation systems

Focus | Navigation systems 94 benefits over simply combining a GNSS receiver and inertial sensor from different suppliers and integrating them manually, through the autopilot or otherwise. Part of the value of these ‘all-in-one’ systems is that they typically provide better SWaP parameters than a user- derived solution. By developing and integrating their own systems, GNSS- INS suppliers can arrange for the RF front ends, accelerometers, gyroscopes, processors and other components to be configured in the tightest space possible, using as little excess material as possible in the PCBs and their housings to reduce weight and cost. Also, by developing all the hardware from scratch, navigation system suppliers could find it easier to develop and fine- tune a Kalman filter algorithm that fully exploits the GNSS and INS to output the most accurate attitude and position data possible. A UAV manufacturer attempting their own base-level filtering, or relying on their autopilot filtering, may not yield the best readings themselves. An integrated Kalman filter will allow accelerometer and gyroscope biases to be tracked, giving closer accounting for their measurements and inherent errors. That gives more accurate roll, pitch, attitude and heading information. And while the GNSS receiver is updating position data at rates of between 5 and 20 Hz, the IMU can produce its position estimates at 100- 500 Hz. Thus, in addition to outputting position at a much faster rate, the IMU can compensate for a loss of GNSS signal much quicker, greatly reducing degradation of navigation performance compared with a system that does not exploit its inertial sensor in this way. Naturally, that could have a range of applications. Autonomous road vehicles for example will experience intermittent satellite signal losses as they drive between trees and buildings, meriting quick compensation from their IMUs. Meanwhile, missions in extreme latitudes can often lose sight of satellites below the horizon, so similar INS integration will be needed to make up the shortfall in GNSS updates. And as integrated GNSS-INS solutions give update rates of up to 500 Hz, this high frequency of heading and attitude data outputs enables the stabilisation of cameras for long- distance tracking of targets, for example in security-related missions. It also means more precise photogrammetry can be taken at greater distances or higher altitudes, capturing larger swathes of land with each shot yet still doing so at the sub-degree and milliradian levels of accuracy. Transponders As the number of privately owned UAVs increases, questions continue to arise over how they can be safely integrated into national airspaces in the absence of comprehensive laws and regulatory standards. For UAS developers or end-users eyeing operations in congested environments, the solution increasingly is to take advantage of automatic dependent surveillance-broadcast (ADS-B) transponder systems that periodically output their location (typically over a 1090 MHz-centred frequency band), enabling others to track them. Transponders designed according to the ADS-B Out standard rely on being able to track GNSS (or specifically GPS) signals. However, a typical GNSS receiver system is not sufficient for simply plugging and playing to drive a transponder, as it lacks a few key technical capabilities for ensuring it is operating within margins of safety. In addition to improvements in latency or static and dynamic accuracy compared with typical UAS GNSS receivers, general aviation-grade GNSS units largely operate with horizontal protection levels that claim with 99.99% confidence that the UAV will never be outside a given radius of the output GNSS coordinates – 50 m, 500 m or 5 km for example. They are also programmed to calculate a horizontal figure of merit, a similar index that claims 95% confidence that the aircraft will never be further than another, wider radius than that of the 99% confidence index around where the GNSS is placing it. October/November 2018 | Unmanned Systems Technology Unmanned defence and security vehicles can depend on dual-antenna GNSS-INS units for long-distance tracking of moving targets (Courtesy of SBG Systems) Growing miniaturisation of GNSS-INS components is enabling more accurate georeferencing in sUAS photogrammetry (Courtesy of Inertial Sense)