autonomous surface vessel

I usually controlling the boat with a RC controller. But in that video I was testing an autonomous approach. It uses a pixhawk controller. It has an IMU and a GPS for waypoint navigation.
The battery on that boat allow 1 hour of function, as a research boat missions are shorter than that.

 

Ok, today I've take all components and weight them.
3D printed Hull: 700 grs
fiberglass: 90grs/mts2 I will need 0.5 mts2 on each hull -> 45grs
epoxy resin: same weight as fiber: 45grs
T100 thruster: 270
So each hull will weight 1060 grs
Electronics: 2300 grs
Echosounder: 700 grs
LiPo battery: 510 grs
Aluminum arm for join two hull together: 300 grs

So I have 1060 x2 + 2300 +700 +510 x2 +300 = 6.44 Kg.
Calculated displacement was 5.6 Kg.
Is it ok to scale up 30% my current drawing in order to get 7.28Kg? or there is not a linear proportionality?
Is there another best option to make my specifications?

 

Sorry, I cannot help with the scale up question, sometimes boats are not scalable to keep the same performance characteristics.

I did not realize you were only running the boat for 1 hour at a time and manually charging instead of using solar cells. That changes a lot of things, it may be possible to achieve 3 meters per second, and efficiency is not as important. If you can choose the time to deploy your boat to when the water remains calm from wind and waves, a keel or special stability may not be needed.

Curious how are you turning the boat at each stopping point? Is the blue robotics motor itself turning or are you using a Rudder arrangement with the blue robotics motor in a fixed mount?

 

Existing displacement is roughly 5.6kg.

To scale, it is by the cube of the scale factor.

So if you increase the scale by 10% you get 5.6 x 1.1^3 = 7.45kg, so this satisfies your objective.

What this means is the all the dimensions, length, beam, and draft scale at exactly the same rate.
It is an exact replica, just 10% larger in all 3 directions.
So your LOA is 750mm this is now 750 x 1.1 = 825mm... and so on.

 

I didn't see an answer to this so here is mine. The thruster position off the back will be difficult to balance or 'trim'. Before you are done you will need to make the center of mass the same location as the center of buoyancy (correct trim) so hanging heavy thrusters off the back will force heavy electronics forward. I also don't see the thruster displacement accounted for -you need the weight of the thruster when immersed in water for your calculations. Having weight toward the ends makes turning more difficult and inaccurate.
If you move the thrusters forward toward the turning center you will minimize the energy to turn and the error (the distance traveled by each hull will be closer to prop pitch*turns). You still want the thrusters aft of center so when they stop the boat coasts straight and predictably and having less hull in prop wash is better. Are the thrusters counterrotating?

 

Elgarbe,
A monohull with the battery and motor in the keel may be a better choice, depending on the application.
What does this ASV need to do?
What does the SOR (Statement Of Requirements) call for?
Is the project just for fun or is there work to be done?
Why autonomous?
Have you looked at other, successful designs?
This has been done before you know.
This may be of interest to you:Dan Hook: Five key steps for the operation of autonomous vessels - The Shipowners’ Club https://www.shipownersclub.com/dan-hook-five-key-steps-operation-autonomous-vessels/
And this:MOST AUTONOMOUS VESSELS LIMTIED AUTONAUT ASV https://www.bluebird-electric.net/artificial_intelligence_autonomous_robotics/most_autonomous_vessels_limited_autonaut_usv.htm

If you scroll down on this page you'll see related posts on this forum.

 

Bluebell has suggested that you use a monohull rather than a multi. He is correct for use in such a project. . The mono is easier and less complex to build and will have sufficient stability. It is more than possible that the mono hull will be lighter in weight than a cat that displaces equivalent weight. Moreover there is much more usable space in a mono to accommodate the electronics, batteries, and servos that are needed.

 

I can't help you with the boat design.

But I spent a lot of years working in remote sensing.

Making an autonomous share the water with other vessels and with owned property safely is a nearly impossible proposition.

People have been working on the land problem with cars for decades, and there are still no really good safe solutions. They still make mistakes of various sorts. So do airborne drones.

On water, the problem can be a lot harder, because weather conditions can cause objects to be very difficult to distinguish from waves, spray, etc. Even for humans, fog and heavy precipitation can be a problem. They can also make controlling a vessel very difficult, in ways your programmer might not anticipate.

Of course people aren't completely safe operators either - but expecting any automated device to do as well at pattern recognition as a person is not reasonable at current technology levels. Faster - often yes - but as well - no. That opens up the possibility of doing harm to people and property, as well as opening yourself up to liability, reckless endangerment, and manslaughter issues.

As an example - boats often miss each other, because the operators draw attention to themselves in a variety of ways - e.g., yelling, using horns, flashing lights, mirrors, and radio contact, all of which might be too sophisticated for current machine "intelligence". If your autonomous vessel can't be proven to have done as well at recognizing these signals as a competent human operator might have done, and did as well at avoiding a collision, any reasonable judge or jury would convict you. Also, kids and some adults enjoy fooling machines, and might be inclined to fool or play chicken with your vessel. And it isn't impossible that something your device does or interferes with in international waters could start a war.

There are other legal issues too. E.g., if you happen to video people without their consent, that is sometimes illegal. And if you enter private waters, or interfere with the navigation of other vessels, or fail to yield right of way when you should, or to pay attention to sirens of police vessels, or interfere with protected wildlife, those are all potential issues. The real world is a very complicated place - much more complicated than any current AI or pattern recognition software can handle.

In the military domain, people accept some degree of error, including with autonomous vessels, and military devices sometimes kill people who shouldn't be killed, or destroy property. But the military gets a pass to some extent, on the grounds of national security concerns, which you might not get. There are also passes for things like weather balloons. But unless you get special legal dispensations too, you may be held fully responsible for things you never even considered.

 

mitchgrunes,
The OP lives in Argentina and has said little about his mission or area of operation other than it may be shallow.
Are you in Maryland, USA?

 

Bluebell:
Yes.

You make an interesting point. Most of the people in these forums love boats for recreational purposes, have a commercial interest, or want to live on board as an alternate lifestyle. For the most part, that means having a human pilot. So it is always possible the o.p. actually wants to cause problems. But paranoia should only be taken so far - and it doesn't make much sense for a person seeking to do bad things to post information requests on a public discussion board, unless they don't care about getting caught. The hard part about using technology isn't figuring out ways to do bad things. Any reasonably bright kid can do that, and it happens a lot of the time by accident. The challenging part is figuring out how to try to make sure nothing bad happens.

 

I'm sorry for a late response, I didn't receive email notification for all this new answer.

Ok, thank for the information.
I was working a a little more with payload and 3D printed a prototype in order to check construction issues.
BTW I've never zed but I'm intended to 3D print the model and fiberglass it. It's an easy way of construct for me.
I will not need two lipo battery. For one hour of mission one battery is good enough. Missions will be navigate some waypoints in an area of 100 mts on a shallow water like the video.
After printed a hull weight 500 grs. And finally, I can reduce some electronics and my weight could be:

3D printed Hull: 500 grs
fiberglass: 90grs/mts2 I will need 0.5 mts2 on each hull -> 45grs
epoxy resin: same weight as fiber: 45grs
T100 thruster: 270
So each hull will weight 860 grs
Electronics: 1500 grs
Echosounder: 700 grs
LiPo battery: 510 grs
Aluminum arm for join two hull together: 300 grs
So I have 860 x2 + 1500 +700 +510 +300 = 4.73 Kg.
Calculated displacement was 5.6 Kg.

I try ti keep current dimension because it will be easy to transport on my car.

So, I will give it a try and keep update the post with the advances.

Thank a lot!

 

Thank you very much, as you can see on previous picture I've moved thruster forward. Your explanation was really clear!

 

The problem of monohull, for me, is the size. I want to design on solidworks, then split on some parts and finally 3d print them, join with glue and fiberglass. A monohull will be too big for my printer and I can imagine how many will I need to split.
As I stated before, this is a research project. I working on coordinated control algorithm for unmanned aerial vehicle with unmanned surface vehicle. So I need some small vessel. I'm working on this topic for last 4 years, so, yes, I already know other successful project. But no one of them are opensource so there is no solidworks or other source file available. There are a lot of commercial vessel, but I don't have the funds to afford one of them.

 

Thank for the comment. I don't expect to share the water with other vessels. I usually work on a place (like the video I've shared) where there is no activities on work days. There is people just in weekend.
The rest of your comment don't apply to my use-case, but thank a lot for the advice's!

 

Print the monohull in two pieces.
Hugely doable and way better in so many ways.
But, you're on your way.
Post how you make out with good quality video.
Cheers!