I wanted to share our latest ROV for ship hull cleaning equipped with a water jet. Last month, we tested it on a ship and got some great results.
we developed a docking and hybrid crawler mode. When the ROV reaches the hull, it switches to a vertical position to stick to the surface before starting the cleaning process. Once attached, the hybrid mode between the crawler motor and thrusters takes over, allowing the ROV to move under the ship and perform cleaning easily. Right now, we’re working on a lawn-mower motion to automate the process.
I’m still looking for the best sensors and ideas to fully automate the cleaning path. If anyone has experience with similar automation, I’d love to hear your thoughts!
I’ve attached a video showing how it works, including the docking and cleaning process, along with before-and-after shots of the cleaned surface. Let me know what you think!
Really Nice Job Ryan - dont have any experience using sensors for this but do have experience in using ROVs to clean hulls.
I guess you could get a different sound reflection off a dirty hull relative to clean so some kind of short range sonar sensitive to thoses differences in sound reflection seems possible in theory at least. Another way might be video analysis that detects the colour of a cleaned hull. Maybe a combination of both would be more accurate but would obviously need to be calibrated for each hull. Limiting factor would be the environment under a hull, particularly when cleaning which would mean sensor signals would be pretty noisy/unreliable.
Have seen some great work developing automated hull cleaning (google Hullbot) but do wonder even with well developed senors/software whether pretty constant monitoring by a human would ever become redundant?
Does your sytem only use water pressure to clean or do you use brushes as well?
Thanks! The automation is meant to assist the pilot rather than make the system fully autonomous.
There are a lot of challenges when cleaning a dirty hull. I don’t think a camera-based system would help much since the onboard camera gets blurry during cleaning due to all the debris in the water.
Our crawler motor has an encoder, so for now, I’m considering implementing a simple motion pattern, like a rectangular cleaning path, but the main concern is motor drift and slip.
We use both a customized ardusub and a crawler control system simultaneously during crawler mode. USBL is not suitable for this type of cleaning automation, so im not considering it.
Hi! Right now, our main focus is optimizing the cleaning process but biofoul containment is definitely something we’re considering for the future. Have you worked with containment methods before? I’d love to hear any insights!
I am in the process of starting a hull cleaning company. Tuning in the rov now. Having issues holding attitude. I’m using hi speed brushes. I continually have software issues. I just restarted the project this money. Been on the back burner a few years. Of course everything I have needs to be updated already.
Wow - that’s quite a large and heavy looking ROV! All the buoyancy foam likely makes it float or neutral, but the large mass may be the reason the control system is having trouble holding attitudes / orientations - insufficient force available!
