Hi Gang, I'm a new guy here with a good background in electronics, computers, and RC vehicles. I have a lot of reading to do, but wanted to start off by introducing myself and giving a general overview of my project.
I work for a company that searches for historic shipwrecks (hoping to find Spanish Galleons loaded with silver and gold). Over the years, I have ran into many situations where traditional survey methods (boat-towed instruments) weren't feasible. Examples here are a small fresh water lake containing a historic WWII aircraft crash, but didn't allow motorized vehicles on the water. I built a small RC boat to haul around my Humminbird sonar unit and located the aircraft that way, Often times the shifting sand bars which originally caused the ship to wreck are too shallow to run a boat over today, even though the area may have been 15-30 feet deep 500 years ago.
We generally tow large and heavy magnetometers and sonar units to locate shipwrecks. I have access to a much smaller, lighter magnetometer (measures minute changes in the earth's magnetic field caused by iron objects like cannons and anchors) which can easily be flown under a traditional RC aircraft. Generally a shipwreck would consist of a large pile of cannons and several anchors, Due to the shoals and sand bars in my area of interest, no survey has ever been conducted here and many shipwrecks are known to have occurred in the area.
The specs are likely remedial for this forum, an RC plane that can fly around 20 MPH hauling a few pounds of extra gear. The autonomous part would need to fly the drone over a pre-programmed course of lanes spaced about 30 feet apart in a lawn-moving fashion. After it finishes it's survey grid it could fly to a meeting point and I could land it manually. Floats or a seaplane are pretty much required, have been toying with the ideas of a large Telemaster on floats or the new 72" (?) drone from Nitroplanes.com
There are areas of dry sand banks and small islands that need to be covered as well, so a boat would not be a good solution. Our mag can see the types of iron deposits we are looking for from better than 50 feet, so if I can fly 30 foot lanes at 10 feet AGL under 20 miles per hour I will see it if it is there.
As i said, I have a lot of reading to do on autonomous guidance systems, but if anyone feels benevolent and would like to help me start fleshing out what will be required, I'd be very appreciative. I'm not afraid to do my own homework, but you guys know the merits and pitfalls of the different systems and could likely save me a lot of time.
I'll update this thread as I go too, and thanks in advance for any early advice.
Jason
Replies
Capturing the audio sine wave should be able to be done a couple different ways - a frequency to voltage converter or similarly structured A/D setup should do the trick.
Multirotors can do the job, but would exponentially increase the shielding requirements, weight, etc. A fixed wing with trailing sensors would add additional reduction in interference that might happen to escape from the shielded controls while the aircraft is in flight. One of the advantages to towed array sonar is to be further away from the local noise source (submarine).
Cheers,
Rob
Rob,
After a little more research, the audio that the user hears is actually just varying voltages that are converted to a squeal by an audio board and speaker. If I capture it before the audio board it is still in voltage form, so most of the work has been done for me. I haven't started reading about what it would take to convert those voltages to numeric text strings and save them to a memory card or send them back to the control station. I'm sure this will be a minor thing once we work out all of the other challenges.
The airframe is the biggest issue and we can't even start thinking about it until we know the speed limitation. I have a pretty good idea of the payload weight and plenty of airframes and types of aircraft can handle it. I'm leaning towards a heli or something like a big telemaster at the moment. My biggest fear is that the boss will just want to use the ardupilot system to be an advanced autopilot for the boat. We have autopilots on our vessels, but they are simple "heading-hold" systems and don't communicate with our GPS's. At that point we aren't really innovating anything, just copying what many have already done. We have another sensor package that claims it can detect non-ferrous metals from 40 feet away. It already has GPS and data logging built in so it may be a better sensor package for us. The problem with it is the size of the coi (9 foot by 5 foot PVC frame). That package under a heli could make quick work of the small islands and dry sand bars we want to look at. Too much data and too many options for my brain right now. :-)
Thanks for the input!
Jason - The Ritewing guys have several wings ranging in size that will be perfect for this application... I know them well and am happy to setup an introduction. They are always good at re-designing their existing stuff to meet your need.
Julian
Very nice, thanks! I will look over their options and contact them.
Hi Jason,
Enjoyed your post.
Have you considered the design of an Akranoplan or WIG or Ground Effect aircraft?
I live in Southport we have lots of shallows and shifting sands similar to how you describe and I presume this would be the North Carolina sands you are talking about? Similar sort of coastline.
My project is to build a load carrying WIG as they are perfect in design for the application I want to use - sounding for whelks. The reason for a WIG is simple - it uses ground effect to sustain lift hence any sandbars/shallows will not effect the vehicle so long as they are not too steep which is not a problem in UK in the US I'm not sure. It also can be made with exceptional flotation as a drone and the payload capability is better than conventional aircraft for the power/ weight issue.
I can't find anyone who is building one yet and all the more reason to want to build it for me, it seems ideal for your application too.
My biggest obstacle in the design phase initially is to increase lift at slower speeds by increasing power without adding too much weight this may not be an issue for yourself. How fast can you scan with a magnetometer?
Just a suggestion
Keith
Interesting thought, and one I have not explored. I am still waiting on some data from the manufacturer of the magnetometer. We switched to flux-gate mag which is said to have a much faster frame rate, most mags only sample about 10 times per second. This slow rate needs about 5-6 MPH to ensure complete coverage. I cannot estimate my required speed until I know the frame rate.
I've had plenty of other obstacles to contend with lately so I hadn't gotten back to that part yet. It is possible we will simply use ardupilot to drive our already set up survey boat and then walk the mag over the really shallow areas. The area cannot be disclosed until after our press release goes out, but it is very similar to areas off the outer banks, along the lines of flying pan shoals or diamond shoals. Any point with sand bars extending seaward pretty much matches our situation. Accurate charts of the shifting areas are non-existant. Some of our boat towed hardware comes in at $30K, so bouncing it off reefs or hard bottom is frowned upon. :-)
I will ponder the WIG or GEA tonight, definitely an interesting thought that hadn't occured to me. Thanks!
You may get some ideas about suspending a sensor by contacting Randy, author of this blog post, one of the links on this page.
Thanks for that! I didn't have enough reading to do...:-) Good info, I feel a late night coming on. Muchas Gracias
Interesting project. Actual weight of your sensor package will be a factor in choosing an appropriate UAV. This page, this page, and this page may save you some search time. For information. :-)
-
1
-
2
of 2 Next