Thruster layout


Hi Everyone,

Over the years of working with ROVs, thruster layout has become very important. The use of a lateral thruster to provide sideways movement is very much needed.

Some ROV's have 4 vectored thrusters on the same plane and 1 vertical, others have the same layout as the OpenROV but use a small lateral thruster through the belly of the ROV.

How will the OpenROV be able to move sideways to offset a current?


From my years of working with ROVs I find that doing "useful" observation almost always requires holding the ROV stationary or nearly stationary while in a cross current. I don't think the present 3 thruster configuration is going to do that very well. You rarely can get precisely upstream or downstream of what you want to look at.


Great point. Something we talked a lot about early on. We had 4 thrusters on the earlier design, but the the three thruster model reduced so much cost and complexity that it seemed like the right move.

At this point, I think you'll have to make repeated passes on an object to get continued view.

What do you think about the potential for clusters of ROV's providing multiple angles?


Yes. The hope for this issue is that we can get enough people working on open source ROV software that can translate to other designs. Including a more robust (but likely much more expensive) design.

Also, I've attached an early sketch of the 4 thruster design that prioritized the strafing control.


I am afraid that multiple ROVs with multiple tethers is going to be a real tangle hazard.

I suppose it might be possible to run a real tether to one Master ROV, and have some short range IR or alternate technology tether to nearby Slave ROVs. Navigation of such a cluster to make them fly in formation would be a challenge of its own.


Cluster navigation is really interesting to me because I think it's also really good for maintaining spacial awareness - telepropreoception - if you will. This is why many driving video games position the point of view behind and above the vehicle you're driving. There are also many secondary advantages of driving pairs (or sets) of ROVs together, such as recovering vehicles that get stuck, or avoiding that in the first place. Moreover, I think that it is important to consider the user experience- being able to see an ROV in the environment it is driving if gives context that tells the story about what is going on much better then if you just saw the video feed without the object providing it. This is a really interesting topic, so perhaps we can start a separate thread on it!


I've considered adding a "vertrans" (vertical-translateive) system by adding a fourth thruster and having there be two quasi-vertical thrusters at 45 degree angles, but for simplicity sake, we haven't consecrated on that yet. Not having the ability to crab from side to side is "good enough" and we could add that later. Also, it might be just as easy to just add a panning or side-facing camera.

It would be good to see other people's thoughts on this though!



Hi Eric.

First of all, fantastic project. By far the best micro ROV i have seen so far.

Having the ability to fly the rov lateral is absolutely necessery. Especially in a situation where you have bad visibility, soft seabed or inside a wreck. Having the possibility to fix the camera on a fixed spot and move around it combinding lat/fwd/aft and spin is very important.

Vertrans thruster config works best on a round vessel, like a ball (ex: HyBall ) or a tube shaped hull. The use of a lateral thruster through the belly of the ROV is the simplyest way of doing it and would work best with the over all design of the OpenRov.

Adding a panning camera is not the way to go. The reason observation class rov's don't have it is because it dosent realy work. The only practical way of manuvering a camera is with lateral and axial thrust, and spin of course. Pannning cameras is mostly only used on work class rov's for manipulator operations.



Douglas, I saw from another of your posts that you live in Cape Cod. You say here that you have experience with ROVs. Did you by chance work at Woods Hole?


I worked for Woods Hole (WHOI) in the late 1980's for the geology department designing/building/repairing equipment to measure tides, wave energy, sand wave movement, etc. under Dr.s Aubrey, Ewing, and Von Hertzen. Most of my ROV experience was with Imetrix (now defunct, Talon, LampRay), Benthos (now Teledyne, Stingray), Webb Research (also now Teledyne, Slocum) and Seascan (Hydroglider).


Is this useful? .


That looks quite good. It may be a bit bulky for the thrust it generates, but it is worth exploring.



I'm not trying to be a smart-arse, but could you please elaborate. I'm trying to sell this thing, and I need all the advice I can get! Thanks. David Young


With small ROVs like the Open ROV it can be difficult fitting thrusters of usable power in the available space. You can replace several thrusters with one which is nice, but it looks somewhat bulky. Maybe a hybrid ROV with one conventional thruster going straight back to provide main propulsion, and two vectored thrusters going port & starboard for maneuvering would be good?

Can you measure the thrust you get? Compare it to a good thruster like a Tecnadyne 260 ( Thrust per Watt and thrust per cubic inch would be good numbers to have.

I like the looks of your 2 thruster ROV. Especially the simplicity of the frame. I guess a camera would go between the floats?

A fun, maybe very simple, proof of concept ROV might be made with two of your vectored nozzles on a frame "powered" by two garden hoses.




Your hybrid ROV idea is definitely worth consideration.

The overhangs on the 2-thruster ROV are designed for mounting (and protection) of cameras and the like, but a camera could be mounted between the floats as you say.