Hi, I am trying to design a 6-DOF ROV that needs to be able to keep its position in high currents (4 knots worth case), is able to face an object from different angles, and can surge forward decently fast (say ~2-3 knots on average, also against currents). This means I’ll also need to be able to strafe and adjust the pitch of the ROV. Another constraint is to build this with as few thrusters as possible, preferably no more than 6 (7 worst case), and I can’t mount any thrusters on the front, nor the back (at least, not aligned with the centerline).
My main topic of discussion here is thruster positioning… I’m very new to this and have had a look at a few examples, and none really seem to do what I’m looking for. I’ve been playing around a bit and thought of this thruster layout (see images - the right side of the screen is the front of the ROV)
Essentially, there’s 2 vertical thrusters that’ll be aligned with the CoG. Two ‘horizontal’ thrusters at the back are rotated 45 degrees inward, then 20 degrees downward. The two ‘horizontal’ thrusters at the front are also rotated 45 degrees inward but 20 degrees upward.
This should in theory allow for surge, sway, yaw, heave, roll, and pitch, right?
In the absence of decent documentation for determining thrust factors, this thread (and others with the vehicle-design tag) may be of interest for some of the factors that should be considered.
Yes - you’re covering all the directions
That said, there’ll understandably be some different control authorities (e.g. less pitch than yaw, more forward than vertical, etc), and there’ll likely be some coupling between some of those axes (particularly without accounting for asymmetric thrust).
I’d think that for a vehicle that large, with that high of a target speed, you wouldn’t want a thruster at all! The nozzle on a T200 or T500, or any thruster, is designed to amplify the force it can produce - you can think of it like a tractor. When you try to go fast, that nozzle actually is a lot of drag and can slow you down! So for ROVs that often need to push hard, great, but for an AUV shaped ROV like you’ve shown that needs to hit those kinds of speeds, only one or two motors with a large, high-speed prop seem better suited?
Hi @EliotBR
Thank you for the reply, I understand there’ll be different control authorities though I think that’s acceptable. The thread you linked looks interesting, I’ll have a look at it.
The length of the tube in images is 620mm long, 110mm OD. I feel like that’s pretty small for an ROV, no? I get where you’re coming from (I’m also a bit concerned about the drag), but I also need to be able to stabilise (stay in place) properly in high currents. I’m not sure how that would be done with most AUV configurations (e.g. 1 or 2 thrusters and assuming fins for turning). The tube is mainly for ease of assembly and because it’s much easier to waterproof reliably than custom parts.
I’ve had a better look at the thread you mentioned with the FIFISH ROV - shouldn’t the thruster layout that it uses work exactly for my use case? Seems like a more logical thruster layout than I initially proposed, that’s easier to compute the thruster matrix for, and I don’t see any downsides yet…
It is indeed another 6 thruster configuration with 6DoF control, if that’s what you’re asking. That said, it makes some different control authority trade-offs (as a couple of examples, yours prioritises forwards authority more, but at lower efficiency, and vertical less, but at higher efficiency). Whether either approach works for your use case depends on your specific requirements (and your hardware).
In the 6x6 case the thrust factors matrix can be quite readily determined by inverting the control authorities matrix (followed by some per-axis or full matrix scaling to prioritise between maximum axis control authority and efficiency), so I’m not sure either frame is easier or harder than the other to compute?
