The National Society of Professional Surveyors has, as its first pledge of a professional land surveyor “To give the utmost of performance”. While this mantra provides a continued mentorship to the professional as they carry out their geo-scrutiny it also acts as a guiding principal for those of us behind the scenes that deliver the tools, training and support of technology.
There are many known uncertainty contributors that make up acoustically derived soundings. If the manufacturer holistically considers these and themselves pledge to reduce these insofar as their business allows then our survey industry gains not only better data but increased accessibility as the barrier to new users is reduced.
The Subsea team at NORBIT (Trondheim, Norway) have grasped this focus wherein all engineering decisions, through the development of their family of systems, have culminated in intuitive shallow water mapping suites deployable in hours, not days, onto any surface platform and operated by the novice.
Surely, some readers will recall the art and science of sonar tuning. The careful balancing of sonar transmit power versus receiver gain. The tweak (taking years to make intuitive) brings comfort to the operator and earns great respect from the recruit, but tantalizing fear on being left alone to manage the system. We now see these manipulations as subjective sonar tuning that brings the threat of measuring too small a signal to make much use of, or over-saturating the returned echo resulting in increased depth determination uncertainty. Recall that receiver gain only increases the already received signal and therefore will do nothing for the Signal to Noise Ratio (SNR). It shall be used to adjust the signal to prevent the saturation or numerical errors, but it should be done automatically and internally to the sonar without any user intervention.
The key is outputting sufficient energy and keeping the noises as low as possible. More energy going out translates to more energy on return. The signal must transit to the bottom while suffering absorption and spreading losses and possibly scattering from reflectors in its path (weeds, fish, suspended sediments, bubbles, plastic bags, etc.). If the transmission even makes it to the bottom, it now is absorbed and scattered with bottom interaction before sending a small amount of signal along the treacherous journey home. In most modern systems today, there is no need to lower the output power as the intelligence in the sonar prevents the hardware saturation. Only in the rare environment (lock chamber, dry dock) might it need to be reduced.
To map to deeper waters or for a wider swath width, there is a need to increase the transmitted energy. This is done by an increase in pulse duration, similar to extending a vowel sound when speaking at a given volume. In Continuous Wave (CW) systems, the increase in pulse duration comes with unwelcome deterioration of resolution which lowers the quality of the bathymetry returns. Manufacturers will offer very short pulse length options on specification sheets to indicate high resolution. But in real life, these values are not useful in a typical environment due to low transmitted energy and poor performance. Give it a try on your CW system and you will see that increasing the pulse duration (TxPulse, Pulse Width, Pulse Length, etc.) will measurably erode resolution while too small of a pulse length will result in a very narrow swath in shallow water. Therefore, in CW systems, one cannot maintain both high resolution and full range performance.
What would be the solution then? Use a pulse compression technique. Pulse compression has been used for decades in radars as they allow use of much longer pulses over a wide frequency range without loss in resolution while keeping long detection ranges. This is possible by employing a long FM signal, called chirp, which sweeps through the frequency range, e.g. 360kHz to 440kHz. The long duration of the signal facilitates the wide swath and the high bandwidth to maintain high resolution.
When working with FM systems, the resolution becomes independent of the transmitted pulse duration as it depends solely on the bandwidth of the signal. The higher the bandwidth the better the resolution. For NORBIT FM multibeam systems, even for water depths of 100m, the bandwidth of 80kHz is used and the sonar range resolution remains 9mm.
FM chirps and pulse compression techniques has been in use in radars for decades and, when correctly employed in multibeam systems, allows for large transmit energy resulting in increased SNR for each beam for reliable sounding determination (cleaner, wider swath of high resolution repeatably measured depths). Shallow water FM for multibeam systems is current state-of-the-art technology and NORBIT systems have been doing this for nearly a decade with many hundreds of satisfied customers globally.
The result is a system that requires, for 99% of applications, minimal manual tuning and minimal user interaction. Simply set the viewing area (swath angle and upper/lower gates) so that the full bottom lies within the maximal range and then turn off the settings tabs and focus on survey management (sound speed, GNSS corrections, line driving for required coverage, avoiding crab pot buoys, etc.)
All beamforming multibeam systems today require the input of a timely local sound speed for correct calculation of beam steering angles. Sound speed measurement errors may be due to many factors such as incorrect probe placement, delays from communications or filtering, probes that are out of calibration, etc. The errors resulting from incorrectly applied sound speed have consequences for the determined beam angle for each steered beam.
To understand the role of surface sound speed on beam steering angles we must know the relationship. The system must determine individual ceramic element time delays for a desired steering angle given an array length (distance between the group of ceramics employed) and to do this, the speed of sound is required (consider that time = distance / speed). The formula is something like:
It goes to show that if the speed of sound differs from known, this error will multiply by the time delay and steer the beams in a wrong direction. The more time delay is needed the bigger error is observed.
If we are able to reduce the needed time delays we are able to reduce the beam steering errors. This can only be done by changing the shape of the receiver array so it faces the direction of the incoming sound wave. A curved array faces the returning sound from all directions and does not require large beam steering as in flat arrays. Therefore, the beams are not impacted by small variations in surface sound speed. A helpful tool, AMUST by DELFT University, available from Rijkswaterstaat of The Netherlands, will provide insights to the effects of accumulated uncertainty and the effects of incorrect sound speed for beam steering and the resulting uncertainty for depth determination.
NORBIT systems always include a surface sound speed probe tightly integrated into the receiver and sport curved receiver arrays. Therefore, the impact of identical error in local sound speed at a 50° angle with flat array will cause the same beam pointing error as one at 80° in a NORBIT array. That is because NORBIT steering reference angle is roughly 30deg. Another benefit is the ability to map with 180°+ wide swaths in shallow water (bank to bank) from a single sonar head or scan to either port or starboard shoreline without physical head rotation.
The sonar is just a part of the systems kit where each sensor works in concert keeping time and pace. Large errors typically originate from mis-alignment of the Inertial Navigation System (INS) / Motion Reference Unit (MRU) axis frame and the offset measurement frame where the longer the lever arm distances are along each misaligned axis, the greater the error. A vessel with an INS/MRU situated in the belly of the vessel may be well placed for heave determination but if slightly (~0.3°) out of alignment with the vessel frame, which is often used as the offset measurement frame, will be sufficient enough to exceed the ability to meet IHO Special Order and even Order 1 surveys in 10m water depth. Especially with a multibeam sonar mounted to the side of the vessel or some distance away from the navigation center, errors will appear and vary in magnitude with increasing vessel dynamics. The solution is to reduce the separations between sensors.
NORBIT has led the pioneering effort of tightly coupling leading GNSS/INS systems into the sonar frame resulting in a setup with non-existent flexing or movement between the sonar and the INS. The lever arm distances are fixed and known to the sub-millimeter. The GNSS/INS is auto configured to output its solution to the same location as the sonar measurement center. The operator need only measure to the primary antenna and offset the sonar/INS shared measurement location from the best-guesstimated vessel center of rotation moved to waterline while the INS reads a near-zero pitch and roll value. The system is now ready for ellipsoidal referenced surveys or surveys using determined waterlevel (tide) information.
A complete mapping system (with optional LiDAR or Sound Speed Profiler) is now included in a single wheeled hard-case that may be checked onto any commercial airline. Only a laptop is missing for a complete system suite.
Systems installation often leads to exhaustion especially when carried out in remote locations and, or when under time pressure and, or when a previously unknown vessel of opportunity is employed. Compromises are often made when deciding sensor locations due to length or path for cables or attempting to maintain alignments or what available hardware to use to brace the system for reduced flexing or vibrations. The larger and heavier the complete system is the greater the size and weight of the mounting hardware. The accumulation of ‘making the best of it’ decisions often leads, at best, to borderline acceptable data.
With luck, the surveyor is now content that the system is installed as best as they are able given the hardware and tools available and must now carry out the survey. This will still not be day 1. It will be day 2 or day 3, at best. If using the NORBIT integrated multibeam system with their Portus Pole, we are only 1/4 to ½ through day number one. Indeed, the complete integrated NORBIT system requires only four bolts to attach the coupled wet-end (sonar, sound speed probe, IMU, fairing and bracket) to bottom of vessel or pole mount, one deck cable from this coupled wet-end to the water-tight topside unit, one antenna cable for each primary and secondary antenna, a 12-28VDC power cable and one Ethernet cable to a PC or Linux machine or laptop. No timing cables, no PPS connections or splitters, no wrong gender null modems, no urgent trips to an under-stocked and now non-existent Radio Shack.
Rapid mobilizations are enhanced with the new NORBIT Portus Pole, a streamlined, no tools carbon fiber pole mounting setup with telescopic antenna mast that packs into a single wheeled case and may be checked as airline baggage. The ability to fix all offsets as well as the multibeam sonar alignment calibration angles (MAC) allows rapid system setup and immediate operation for repeatable high resolution multibeam data.
NORBIT will take care of the millimeters/centidegrees and the quality high definition and repeatable survey will take care of itself.
Once afloat with systems installed, the surveyor must now configure the data acquisition system. This requires setup of sensor device drivers, communication protocols, offsets between sensors and the Center of Rotation and project coordinate reference system relationships with respect to the GNSS ellipsoid. Then, the user must configure and upload background imagery, build display databases/grids for bathymetry data to be collected, setup vessel tracking point and arrange various windows on whatever screen real estate is available to them.
The need for multiple data windows during survey acquisition has historical merit and depending on the survey purpose, may still be required for a very small select number of surveys being conducted today. However, for the majority of surveys (especially those that utilize real-time GNSS corrections for a 3D positioning solution or for post processed positioning) most of the display screens are not necessary except for those who have a determined interest in real-time angular rates of acceleration or velocities. There can be well over 200 items showing real time quality. What is required and what just confuses the operator?
NORBIT is the first (and still the only) company to have fully integrated a GNSS/INS system for bathymetric mapping. Shoving GNSS cards in a sonar topside is relatively simple. The full system (sonar, GNSS and INS) are monitored and controlled from a single interface and connected without additional cables except to the antennas. Raw GNSS/INS data is recorded automatically for delayed-time heave, for PPP or PPK. Now, let’s move this to the left side of our single laptop monitor and open up DCT, shall we? Data Collection Tool is a new NORBIT browser-based data acquisition platform with full open access API and powered by current geospatial imaging techniques. Google satellite view or Open Street Maps backgrounds are available online (with internet connection) or for may be downloaded for offline use. Mouse-controlled operation allows one to acquire ultra-high-resolution data, or, use your finger on another touchscreen tablet or smart phone. All data is recorded when the recording button is pressed with the file name displayed beside it. Throughout the survey, any critical issues are set to alert the operator (timing synchronization, surface sound speed issues, poor positioning quality, loss of GNSS corrections, poor bottom detection quality, etc.). It conceivably cannot get much simpler (but, my crystal ball says that it will).
As I write this article on behalf of NORBIT, I ponder what the old salts (my career mentors) might complain about the direction that NORBIT goes and if they might proclaim the needs for separation of sonar from INS, that FM doesn’t work for shallow acoustic systems, that curved arrays are yesterday’s systems, of the need for 22 separate windows to be visible on an acquisition system, on placing the IMU/INS at the vessel COR or perhaps my bad jokes. I say, sorry, but this is progress and currently, this is NORBIT. New blood, new thinking, new energy, fresh pioneering.
The wild success at NORBIT comes from the pure potential of the enthusiastic open-minded team. The genuine desire to engineer practical and efficient solutions. Part of the success comes from NORBIT seeking out groups within the industry to understand the challenges of our survey industry and to build lasting partnerships with. Seahorse Geomatics is one of those partners. It’s telling that the fresh pioneering implementations engineered by NORBIT are taking hold throughout the industry. FM, curved array, tight integration of GNSS/INS, simple user interfaces and immediate access to repeatable clean data. As NORBIT continues along the path of reducing the sum of most potential sources of error we can say that “a rising tide floats all ships”.
Background in Geomatics Engineering and worked both as a land surveyor, hydrographer and in multibeam R&D since 1998, Mike Mutschler founded Seahorse Geomatics (2011) with Heidi Seger to serve surveyors and those in product development. Bathymetric multibeam courses are held twice annually while on-site training is delivered to navies and Hydrographic Offices worldwide. He has partnered with NORBIT AS since 2012 and is, at the time of this writing, aboard the R&D vessel “SheHorse” on the Columbia River, testing the new WINGHHEAD i77h, a 0.5° x 0.9° (400kHz) ultra-high-resolution mapping beast with Dr. Pawel Pocwiardowski remotely connected from Santa Barbara.
When using dense, high-precision survey data, the method for management and visualization of the data can have a large impact on the final decision making process. This is an important factor when accurate shape reconstruction is required, as there are significant trade-offs with traditional approaches. For applications where it is critical to know exactly the shape and size of surveyed objects, a high-resolution 3D mesh is likely the best option.
There is no doubt that sonar has revolutionized the way in which we are able to map the seafloor. This huge advancement, however came with one key drawback — the introduction of ‘noise’.