Bottom line up front: GPS works on ArduSub and BlueROV2, but it needs to be mounted on a mast and out of the water to work properly.
After getting a proper tablet earlier in the week, I set about doing various ROV tests. One of these was to finally get the system wet.
The results are below.
Screenshot 8 shows the ROV out of the water with a good GPS lock.
Screenshot 9 shows the ROV in the water, disarmed and on the surface. You can see how it looses lock and the GPS wanders according to inputs from the IMU.
So what next? Well, we are either going to need to make a mast and mount an antenna externally or reconfigure an antenna inside the hull. There isn’t a lot of room to do the latter, so I am open to ideas.
Oh, I see now with the screenshots how you see that the gps is working/not working. I test with a usb gps on a mast. Do you have a raspberry pi on your ROV? I can explain how to get the gps into ardusub if that is the case.
Ok, first of all, you will be limited to about 3ft depth, or whatever your mast is, but I wouldn’t recommend a longer mast than that (make sure your mast is stiff!). I use a BU353-S4 gps, and scripts designed specifically for that unit. The first step will cutting the usb cable so that you can penetrate it, then reconnecting the usb once penetrated. The usb gps plugs into the raspberry pi. I will provide some more details at a later date. I will update the ArduSub documentation in the meantime so everyone can benefit.
Thanks, I’m familiar with the BU353-S4. I’ll order a few and get one integrated with a penetrator. any idea if the GPS is sealed, or does it need to be encased in epoxy?
I’m thinking of making a 3D printed mount and using some of the existing mount holes. Be a nice little addition.
I can confirm the same gps behavior. It seems like anything more than a foot underwater and I lose gps. It’s still handy to have once you are surfaced.
I mine is a custom frame but similar layout to BROV2.
Thanks guys. I tried to design it to be as compact as possible using some left over materials. I’m still not 100% satisfied with it and would like to add clamps for the center tubes.
It has been in the water. The first test was in a quarry here in KY. It made it to 40ft with no issues! Credit goes to everyone that has been developing ardusub.
Good test Kevin, although the results were pretty much what a lot of us thought would happen. FYI, the VideoRay CoPilot system uses an active GPS antenna potted in a shark-fin / surfboard fin apendage on top of the float block. I rather like that idea over a wand, less likely to snag.
The Riptide UUV has a similar setup with the GPS mounted in a fin. The fin works fine for vehicles that go straight, but we have strafing capabilities. I’m thinking of more of a “support” that is mounted in a few of those holes on the top of the frame. So not a wand-like mast, but maybe something angled that gets the sensor up about 2 inches out of the water.
Yes, that is to be expected. But did it work well while the ROV was on the surface? I plan on building a waterproof “pod” and “mast” to hold the GPS above the water’s surface.