Unmanned Systems Technology Dec/Jan 2020 | Phoenix UAS | Sonar focus | Construction insight | InterGeo 2019 | Supacat ATMP | Adelan fuel cell | Oregon tour | DSEI 2019 | Copperstone Helix | Power management focus

40 Sonars for fishing Sonars for commercial fishing represent a new variety of USV payload that is rapidly growing in popularity. The drive for autonomous and persistent fishery surveys comes from the need to reduce ‘bycatch’ – fish that are discarded at the end of a day’s work because they are too young or the wrong species. Fishermen sometimes find more than 50% of their ‘yield’ is bycatch; such waste harms fish stocks and ocean wildlife, and reduces profitability. Fish sonars can be developed around a single-beam echosounder, which will capture the direction of and distance to a school of fish, as well as readings on target strength and additional detail depending on beam width. Alternatively, if a multi-beam echosounder is used, the sonar data from that system can be processed to produce a measurable 3D model – in this case, approximating the school of fish rather than a section of seafloor. Critically, these sonars require high accuracy in their ‘target strength’ variable – the accuracy with which acoustic energy is received after being reflected by the sonar’s ‘target’ – ideally an expected measurement error of no more than 10%, or 0.1 dB. The reason is that biologists and fishery scientists have in recent times optimised formulae linking acoustic target strength measurements with the physical sizes of the fish. That enables processing software to model and report on their maturity, helping a fishing crew decide whether to pursue and catch them or seek a better haul. Also, if the echosounder can be configured to measure the strength of volume backscattering (backscattering within a given volume of sea area), another formula can be applied to measure the density of fish in the school. Not only is that a key measurement of the yield to be gained from the school, it could help discern the size and shape – and thus the species, as well as maturity – of the fish. USBL systems The development of compact high- end acoustic positioning hardware has accelerated to fill the gap between large, expensive high-end UBSL systems and small, poor-quality ones. Moreover, the creation of new USBL architectures from ‘blank sheets’ has spurred the adoption of digital processing electronics to replace older, analogue signal processors. New acoustic positioning transducers are thus less complex mechanically than their predecessors, cheaper to produce and can be packaged smaller than ever, while retaining the same gain levels, and offering wider operating bands. That allows UUVs to be smaller and still have new and complex algorithms embedded for in-situ processing and autonomous behaviours. Of these algorithms, perhaps the newest and most important from the past few years are those that positioning transducers use to calculate the speed of sound underwater, which directly relates to the transmission and reception between hydrophones. Traditionally, the speed of sound has been measured before a mission by surveyors performing ‘dips’ with calibrated instruments, the reading then being loaded into the USBL system. However, the quality and accuracy of the measurements are closely tied to the skill and experience of the surveyor, so errors are often introduced into the range of measurements, headings and angles. In some cases, surveyors have been known to guess at or even forget to enter critical input values needed by the USBL for it to give accurate data. By autonomously accounting for this, errors in reported vehicle positions and headings can be continually minimised in real time, particularly by the widening pool of users who are not hydrographers by trade but owners of subsea and waterside infrastructure seeking to adopt UUVs for regular surveys. By making USBL (and other acoustic systems) more plug-and-play in their design, they can also save money on the surveyors and software licences needed for surveys. Also important is the relatively new ability to calculate transmission and reception angles at both the surface USBL transceiver and the UUV’s onboard computer (or at its transceiver, if one is installed on the vehicle). That enables relative heading changes at the surface and subsea beacons to be calculated, to output an ‘absolute heading’ measurement for the UUV without necessarily needing a FOG for that level of accuracy. This circumvents a key problem affecting many USBL systems, which is their sensitivity to the sea state and attitude of the platform. Accuracy in acoustic positioning has often been inextricably tied to mechanical December/January 2020 | Unmanned Systems Technology Traditional sidescan sonar creates a gap in the centre of the image. By mounting the arrays angularly, the beams extend across the gap ahead of the vehicle, capturing the entire swathe (Courtesy of Klein Marine Systems)