Autonomous ROV - For StarSplitter

[ScubaFinder]

Hey bud...saw your post on the other thread....and thought we should start a topic on it. It's great to communicate via PM, but it helps everyone and gets more input right out here in the open...not to mention it pertains to treasure hunting. LOL By autonomous, I assume you are talking about a pre-programmed route...drop it in and let it do its thing? That stuff is cool, but I'm working on a wireless...controllable ROV that I thought you might find interesting too. There are several colleges working on autonomous vehicles of all sorts...air, land, and sea. I assume you've checked out the Hornet II...there was an AROV on that site as well.

I've been doing a fair amount of reading and thinking on this. I've basically found a solution for cable-less control, but it's not exactly handed to you on a silver platter. I had to call in my electrical engineer buddy, and even he is having to read and think about it a bunch...but we're getting there.

I found the original info on a page about a huge radio controlled submarine (scale German U-boat like 8 feet long). The guy had built a comm box that converted a standard RC signal (from a radio controlled plane transmitter...my other hobby) and converted it into transponder pings (sonar). He used a transducer in the sub to pick the pings up and converts them back to standard RC. Then he feeds them to a standard RC reciever which controlled servos and electronic speed controls for the electric motors.

I don't still have the original link...Google it and I bet you can find it. I printed it all out, and we carried it around and spent several late nights at IHOP downing coffee and trying to figure out a plan. I've since lost the original docs...but have piles of my own now. We did it from scratch, and it wasn't easy. Basically we have a workable plan in place, and my EE buddy has successfully converted the RC transmissions into readably spaced sonar pings. He's working on the receiver now, which in effect will be the easy part now that he can convert and transmit it.

He used PIC boards (google again) and a programming language that I can't recall to do the fun stuff. Basically I'm waiting for him to finish the RX so we can water test it. If it works, he'll draw up schematics for the electronics and I'll write some simple instructions for the programming part. It was really easy...my part...building and setting up the boards was WAAAAAAY beyond my level of expertise.

He got me 8 independant channels of communication. Our only concern with this end of it is whether or not our program will be able to tell the deifference between original pings, and reflections off the bottom. We wrote a really complex filter to take the strongest signal, and only accept singals that were within a certain percent of the strongest strength...in theory, no matter what depth we are working at...it should work....but me thinks we will have a fair amount of tweaking to do in that arena.

The real problem here is in transmitting the video back to the surface...that is a LOT of data...too much to transmit using sonar transponders. We're still looking for the answer there...we have considered sending back a crappy 8 bit 4-color image that will allow us to steer away from large obstructions...all the while the camera is recording top notch video for later study.

Anyhow, maybe that will get you thinking...it's been a fun project...not sure what the heck I'll do with it...I'm a diver...who needs and ROV.

Jason

 

Amazon Forum Fav 👍

DR.ÖTEK Metal Detector, Waterproof Gold Detector, Pinpoint & Terrain Modes, Professional Metal Detector - Grab it through Amazon!

[ScubaFinder]

Thanks Cornelius...I have solved the battery problem by incorporating Lithium Polymer batteries...they are 3.6 volts per cell (4.2v fully charged) and have an insanely high milliamp rating for their size and weight. We have 24 amps of capacity at 110 volts in a very small and lightweight package. Theoretically, we can leave the air in the PVC and establish slightly positive bouyancy with little effort.

I haven't found a good solution for receiving the transponder signals and converting them into readable commands...maybe you could shed some light on this? I used the RC TX and RX because I already understood how they worked...what would you use to RX the transponder signal...I assume we could leave the RC stuff out and use the PIC boards to convert...but I'm not sure how to go about that just yet.

Look into the LiPo batteries mentioned above...outside of expensive fuel cells, they are the best thing yet...AND they are inert in salt water, so if you spring a leak...they will just fizzle out as opposed to shorting out and ruining the entire ROV like Lead Acid batteries.

Jason

 

[ScubaFinder]

StarSplitter, your post came in as I was typing. I like the idea of an autonomous boat too. Perfect, preprogrammed search patterns would be very useful too. No reason why you should be towing a myriad of devices behind it though...I'd go magnetometer, side scan sonar, and sub bottom profiler...all at once (cause we're all made of money right?)

You should really read up on some of the college research projects...the group that did the Hornet ROV has some great stuff in the works. Their vehicles are underwater...basically look like a side scan towfish...but it has a brain. They go frop it off, and it navigates a search pattern underwater. An above water vehicle would be easier and less expensive...but with all due respect...it's have to be less expensive than a good captain...cause I'd rather be ON the search vessel, ready to jump in and investigate any anomolies as they were recorded. More thought needed....LOL.

Now, if you convert this invention to a very small vessell...like an RC boat...you could have some real savings in fuel, slip fees, crew, etc. But you still have to have good GPS to mark the location so you can get back to it.

Jason

 

[rssharpetx]

Jason,
One way we have gottin around acoustic multiples (reflections or ghost, surface reflections) is to use a primary frequency say 12khz , encode a second frequency say 120hz or dual tones like on your phone. Use a phased lock loop to seperate your primary 12khz and dual tones at the receiver. The PIC has built in PWM or analog inputs to decode dual tones. Now assign tones as 1 or 0 or simple commands ,you now have binary scheme but VARY poor baud rate. Typically we use a phased array to $ for homebrew......or just make it simple ... multiples have a phase shift....and you know when you transmitted , lock out sections of time you dont want..........average and look at the slope or rise time of echo.....I have done this with pics before...

Anyway my 2 cents
good luck and best regards

 

[Salvor6]

Hey Starsplitter, here is an AUV for you. Go to: www.ocean-server.com and click on the Iver2 AUV link. It is an awesome machine. The guys that build it use it for bathymetric surveys. They have no idea it could be used to find shipwrecks. If treasure hunters get a hold of this machine, it will be all over.

 

[ScubaFinder]

Thanks RSSharpe, hadn't thought about a phased array..if our simplified version doesn't work I'll give that a try. The more i think about it, a snorkel with a TX/RX combo on top will probably be the final answer, because like Cornelius says...you really need visual.

I'm actually building in effect a "steerable camera" that you just drop down on a cable and tow it along with the boat. It's a towfish with a camera in the dome on top at the front. The camera is mounted on a 360 degree rotatable base, with pan, tilt, and zoom controls all wired up. This doesn't help much in looking for 350 year old wooden ships...but that's not the only projects we are working on.

The other is like a pet project...just something fun to play around with. I want the functionality of a small manned research sub...in a smaller, non-manned, wireless model. I don't ask for much right?

I'm off to check out Pete's link...he peaked my curiousity.

Jason

 

[ScubaFinder]

With that type of power...there's only one way to do it...the lithium polymer batts I mentioned above.

 

[FISHEYE]

i had some ideas on a autonomous about a year ago.just setup a 14 foot boat with a auto helm(i have several)a laptop,GPS,4 stroke outboard motor,or small diesel motor run on FREE veggie oil,side scan or mag,or both.wireless networking,wireless video.i have both,the wireless video i have has a 1 mile line of sight range.i have programmer friends that can write a small program,where that boat will run a grid or north south lines.the robot can be set that if it gets out of range to shut down.anything can be programed into it.why waste fuel in your big boat magging or uw video searching,let the AUB do the work for you.then all you need is a area to search an the time to spend fishing or catching lobsters while your AUB does the work for you,then read the data later.

 

[Salvor6]

My buddy rented a house to the Iver guys while they were demonstrating it to the brass at McDill AFB. They said it has LiPo batteries and when the Iver comes along side your boat at a programmed meeting spot you can download all the info to your laptop via WiFi and upload a new search pattern. They also deployed it from the beach with no boat and after scanning the bay all day, it came right back toi the beach. The Iver sells for under $50.000.