Hello all,
We purchased one from Blue Robotics, we’re still in the process of machining our mount. Just curious of any success stories, it’s hard to find much info on this sensor.
Thanks,
Brad
Hi @rokclimr -
We just helped Ocean Rainforest survey 80 acres in a morning with one!
What sort of information are you looking for? Anything not on the product page, particularly under technical details?
Hi @tony-white
Thank you for responding! We’ll have ours going in the next couple of weeks, I was just hoping to hear about anybody’s success stories/experiences with the unit. There’s just not a ton of user info out there.
Are you feeling good about the data your collecting with the unit? Any issues in using or setup? Are you using with a Blue Boat or custom setup? Ours is going on a custom setup.
Thank you!
Brad
Hi @rokclimr -
What would you consider “user info” - first-person accounts of usage?
The data I’ve collected with the unit in many different tests has always been quite impressive - I even have resolved a structure floating at ~175m depth, in 400m of water! All of our testing has been with BlueBoats only…
The integration guide has been carefully crafted to ensure you have a painless setup and usage experience! Even if you’re using a different boat, as long as you’re running BlueOS and the SonarView extension, or have SonarView running on a reliably connected computer that can reach the sonar’s IP, it should just work!
We’ve sold almost 40 units thus far, not to mention the ones that Cerulean has sold directly to customers - I’ve not heard any performance critiques!
Tony, that’s quite impressive! We should have ours in the water next week, I appreciate your help.
We are currently using the Surveyor on a Blue Boat with great success. We had some initial issues with the unit, but Cerulean was very quick to help and resolve the issues we had. I collect the data using Sonar View and import it into Hydromagic to clean and process.
With a RTK installed the data is very accurate and reliable and similar in quality to data collected by a Norbit system that is also on site. We’ve found it a cost effective and easy to deploy method of getting reliable bathymetric survey data for dredging.
@Randomguy22 that’s great to hear! We’ll be using a RTK setup as well. What are you using for heading data for the Surveyor? Are you using a differential GNSS setup or the stock compass? Thanks for chiming in!
Brad
We’re currently using the stock compass and an emlid M2 unit. After setting it up we did some changes on how much the compass effects heading when it’s moving as we are working in close proximity to barges. We had a few extra Trimble marine antennas so we are currently using that as the antenna for the RTK. With the cellular modem kit and NTRIP extension it follows the survey plan exactly.
When using the Cerulean Surveyor 240-16 on the blueboat through BlueOS, is possible to feed a true heading (HDT) from a directional GNSS (as well as the RTK GGA)?
Hi @Eloi -
Absolutely! To clarify, you’d need to feed those GPS messages to the autopilot, which will use it for navigation, and when queried by SonarView, provide the position and heading information. The BlueBoat FAQ contains instructions on how to set this up:
Any NMEA position stream can be parsed by the Navigator Autopilot. To receive the messages from a network-connected device simply configure the Autopilot parameters GPS_Type to NMEA, Serial2_Protocol to GPS, and verify no other Serial port # protocol is set to GPS. Finally, under Autopilot Firmware, Serial port Configuration, set Serial 2 to: udpin:0.0.0.0:27000. Then direct your UDP NMEA stream to the IP address of the BlueBoat. Note that these NMEA messages are required for navigation: RMC, GGA, VTG, HDT. This approach uses the ArduPilot driver to interpret the NMEA messages - you can also use the NMEA Injector in BlueOS, which has support for fewer messages. If using a serial connection to bring the NMEA position data to the Navigator serial port 3 or 4, set the appropriate Serial# Protocol parameter to GPS and restart the autopilot. NMEA messages should be sent at 10hz or faster.
@Eloi Ours won’t be on a Blueboat, but we will be setting it up with dual RTK GNSS for heading as well. Unfortunately, our unit had a networking issue and had to be returned to Cerulean…hoping to have it back next week.
Outside the BlueOS environment, yes you can still absolutely bring the RTK GNSS heading and position into SonarView.
Hi @Randomguy22 ,
I was really glad to read that you were able to achieve good results with the Surveyor — that’s very encouraging.
We are currently testing a similar setup, using an external GNSS in NRTK mode and, for now, heading from the onboard compass.
On flat bottoms the results look quite good, but we are experiencing some discrepancies on sloped areas, likely related to timing/latency or to pitch/roll/yaw handling.
May I ask if you are using SonarView for data acquisition, or if you are working with other software where you perform patch test and/or post-processing?
Also, in your experience, what level of accuracy/precision are you achieving with your surveys?
Thank you very much in advance for any feedback — it would be really helpful for us.
Currently we are collecting the data in Sonarview and then importing the data into Hydromagic for processing. I have all of the offsets for the boat set up in Hydromagic and it automatically adjusts the data whenever I process the svlogs into a matrix. With this method I haven’t noticed any issues with misallignment of data. We also use a Valeport Swift sound velocity profiler to get our sound velocity adjustements. We then export the data into Beamworx Autoclean for any anomalous points and finally, Hypack, as that is what our excavators/cranes use for their dredging computers.
We have access to all of these softwares for different reasons but none are truly necessary for a good output in my opinion (though they certainly make it a lot easier). We had hydromagic originally for a single beam set up I had made out of a fish finder and gps and then eventually an RTK unit. Dredgepack is a version of Hypack that we use with our dredging equipment. Beamworx I had bought to eventually add lidar mapping into our surveys (Hopefully using the Blueboat).
As for accuracy, on the jobsite we are working with another company that has a norbit multibeam system and we are getting the same xyz results that they are though slightly less detailed on the actual xyz data. For our purposes the data is being interpolated into a gridded matrix so the fine details are less important at the scale we’re working at.
So far I’ve been really pleased with the results we’ve gotten from the surveyor and are even planning on upgrading our Omniscans that are on our Blueboat into the 3d version for even more mapping capabilities. They’ve been really great about answering any questions or issues I’ve had with the unit (mainly due to operator error).
Hi @MatteoBucci ,
We’re working on this very issue (possibly with your company?) at the moment. It appears that the position data coming through via the EKF in ardupilot can effectively appear to have a time delay. It’s on the order of a couple hundred msec, but if you look closely at the data going upwards vs downwards on a fairly steep slope, you can see the position error.
We can now compensate the position for a fixed time delay in SonarView, and we incorporate this feature into a coming release.
Hi @ljlukis
yes, exactly — it is with our company.
Thank you very much for your support. You have been very responsive, collaborative, and extremely helpful in addressing this latency issue. I have already tested the option to compensate for the error on previously acquired data, and it works very well.
@Randomguy22 , thank you as well for sharing these details about your workflow. They are very useful for us as we try to better understand what the best setup could be on our side too.
May I ask you a few additional questions?
-
For the GPS position and heading used by the Cerulean, are you taking them from the BlueBoat EKF output, or have you configured NMEA to go directly into SonarView as described here?
https://docs.ceruleansonar.com/c/sonarview/session-configurations/surveyor-±nmea-heading-position -
How are you logging the data? Are you doing everything through the SonarView extension and saving the SVLOG, then exporting to GSF, or are you using Ping-Python instead?
https://docs.ceruleansonar.com/c/surveyor-240-16/getting-started-with-ping-python -
We do not yet have a sound velocity profiler available, but we are considering adding one to improve the final result. At the moment, we have been working in shallow freshwater, where we know the approximate water temperature and can reasonably assume it is nearly constant. In your opinion, would using an SVP still be worth it in this case? Also, do you know whether Hydromagic allows you to apply a constant sound speed value instead of a full profile? In our case, that would likely be around 1445 m/s rather than the 1500 m/s currently used in SonarView.
-
May I also ask what grid cell size you typically use when interpolating your data?
-
Are you applying any additional motion or attitude corrections in post-processing, or in your workflow are the vessel offsets and sound velocity corrections sufficient?
I fully agree with you — the Cerulean team has been genuinely kind and very helpful in answering questions and solving issues.
Thank you all again for your support and for sharing your experience
I’d love to answer any questions you have!
- We are using the EKF from our blue boat in SonarView. We installed an RTK and the NTRIP extension on BlueOS to get our corrections.
- Originally we were collecting data in SonarView and then exporting a .csv and making it into an XYZ file. Currently we are logging the data using SonarView and importing it into Hydromagic. Hydromagic natively imports svlog data and has saved us a lot of steps.
- I don’t personally think an SVP is a requirement in shallow freshwater if you adjust the sound velocity profile. It does make some difference especially in the conditions we’ve been using it in the winter but not enough for it to be a necessity. Our contract just requires us to take readings from a SVP each time we do a survey. Also before we bought one I had just installed a temperature sensor ( Fast-Response, High Accuracy (±0.1°C) Temperature Sensor ) onto to the Surveyor itself using this guide( Temperature and Pressure Sensors | Surveyor 240-16 | Cerulean Sonar Docs ) and then adjusting the sound velocity manually. This worked well enough for most of our surveys.
- As for grid size it’s dependent on what the data will be used for. In Hydromagic you can adjust the output grid for a matrix when you generate it. If we’re sending it to an excavator or crane for dredging it’ll usually be a 3X3. If we are using it for finding something or as 3d image to show a customer we’ll either load as an xyz or do a .5X.5 grid matrix.
- With the RTK NTRIP set up and the internal imu we haven’t found it necessary to add any extra post processing. I will take the xyz into autoclean to clean up any data that is obviously incorrect and to filter the surveys though this is mainly a time saving measure vs using Hydromagic.If you have any more questions please feel free to ask!
If you have any more questions please feel free to ask!
Has anyone used Hydromagic to capture data directly from the Cerulean Surveyor 240-16? I noticed there is a plugin for this sonar that shows up in the device list.
Would there be any benefit adding an additional IMU to the setup? I have a SBG unit that measures heave that I used on my single beam.
We are currently looking at the surveyor 240-16, and from reading above I see people are getting some good results.
Would anyone be interested in sharing any screen grabs of the raw / processed data. We have used Norbit, baywei, R2 Sonics. But the price point of the surveyor 240-16 really sparks my interest for auv surveys.
Also if anyone has worked in deeper water. We have an upcoming job in mine pit that might be over 120m in areas.
Cheers,
Pete.
Would be great to see other users files, but you can check out the 737 Airliner Aritificial Reef file at sonarview.io. There you can actually run it in SonarView.