Canvassing Engineers to contribute to potential self parking and recovering dingy for yachts. “Dingy, come and get me” “Dingy, go dock on the boat”

[Savannah]

Hello engineers. I’m a retired engineer looking for others who don’t die laughing when I say a self parking dingy shouldn’t be too hard. It would collect a person from shore or drop them off before docking itself back on the yacht. The dingy lives on the yacht.

Initially very much an amateur activity and more for fun than profit. So consider it non-commercial as it needs to be to be on this forum.

I have a pretty good idea of how to do this and what technologies have arisen recently that would make this possible. However, I want to start with an open brainstorming and just see who might be willing to put some time and resources into this and what ideas they have.

There are a few subsystems that could easily be worked on by different people.

Dingy propulsion,control guidance.
Yacht winch control and position.
Dingy to yacht dock system.
Yachtsman remote control/mobile phone
Communications and security

I’m ignoring the political and legislative aspects right now.

I am pretty sure I can do this alone but it would take me around a year and I have other things I want to do. So I think it will require at least 3 people with a range of skills and preferably 5.

There maybe companies already working on this. Maybe companies that would donate electric outboards or something. There maybe a solution somewhere - I’ve not found it.

I live in London and my boat is in canvey island. If you’re interested introduce yourself, your expertise and your location in a message here. Try not to start the brain storming but rather introduce yourself and say how you are willing to contribute or if you are just interested to follow from sidelines. I will do the same and hope that tumble weed doesn’t blow through. :0)

Carl

 

[Savannah]

My name is Carl

I am a chartered engineer. My degree is control engineering. Experienced in embedded systems. I have experience of software and hardware development for realtime systems over many decades.

In 1980 I built the first Vmg meter for a yacht and offerred it to the British Americas cup team. They didn’t want it. The year after this B&G brought out the first commercial instrument system to do this.

I have recently implemented a remote control for an old large yacht to enable single handed docking from low cost components.

I am also able to design mechanical.

Direct and recent experience of the following:
Fusion360
Platformio
Arduino, esp32
Compass modules
3D printing

That’s is probably enough for now. Anyone else who might be interested write something similar length and post here. Thanks. Carl

 

[sarabande]

I suspect that the most difficult part will be to develop the control and navigation algorithms to look after the dinghy in waters that are often heavily populated by vessels that are under human control. Your proposed dinghy will have to address the same challenges that face the Amazon wheeled delivery robots (Scout)

Meet Scout

I think it would be rather unwise to avoid the political and legislative aspects at this stage, as there will almost certainly be essential design features arising from them.

Some serious project scoping is required before you proceed, not least is the commercial viability of such a dinghy, and the ability to conform to COLREGS.

 

[RupertW]

It’s a lovely idea though. We’ve been in a few places where you are allowed to drop off people at a quay but not park your dinghy there (so have to pay 20 or 30 euros for a 5 minute water taxi if you both want to go ashore) and many more places where you have to leave your dinghy on a beach with all the possible security issues. But the legalities feel like a nightmare.

 

[dunedin]

The trickiest bit would probably be finding a way to approach a crowded dinghy dock with lots of boats tied up 2 to 3 deep.

Mind you a dinghy that could move off the dock and hold a geostationary position 50m out until ready to return sounds a nice idea.

 

[ridgy]

A first step would be a remote controlled dinghy so that the user could drop themselves off and then return the dinghy to the boat under their own control.

Bishop was doing it nearly 40 years ago so it should be a piece of piss by now:

An automatic mode could be added on top of this that the user would be free to employ if the circumstances allowed e.g. deserted anchorage.

How does the dinghy reattach itself to the boat when it gets there. I'll give you a starter for 10. An electromagnetic buoy suspended from a cleat that is turned on as the boat approaches and turned off to release it. I guess the magnet only needs to be on the dinghy, the buoy attached to the boat is just a lump of galvanised metal.

27kg pull should do the job:
44,45mm Durchmesser x 41,275mm Dicke Elektromagnet mit 6mm Montagebohrung - 27kg Zugkraft

For propulsion a Bison type thing would do it unless going long distances and cheap. I guess its just a variable resistor in the throttle that controls the speed:
Amazon.co.uk

PyPilot will handle the navigation via a raspberry pi and a small actuator attached to the motor tiller:
pypilot - open source marine autopilot

I don't see a solution to automatically navigating unless you're going to build your own small radar but it would seem a lot of effort when the user can just control it themselves should they need to.

 

[Minerva]

The trickiest bit would probably be finding a way to approach a crowded dinghy dock with lots of boats tied up 2 to 3 deep.

Mind you a dinghy that could move off the dock and hold a geostationary position 50m out until ready to return sounds a nice idea.

Building on this, Perhaps "all" thats needed is to go out a bit then drop anchor? I bet you could use one of the winches attached to 4x4's to let out an anchor rode - just need to ensure the remote control works up to 100m away or so.

 

[Sandy]

I'm also a retired engineer, IET with a Safety Engineering background. You might be interested in some of the research at Plymouth University.

Autonomous Marine Systems Research Group

I am Plymouth based.

An interesting set of challenges. My charge out rate is £.........

 

[Savannah]

Welcome everyone. Some great ideas and points raised in your emails. We have a range of views from the comprehensive to the guerrilla development. No suggestions are too daft at this stage. I would like to gauge how many of you would enjoy contributing directly by making or doing some aspect of the system with a view to having a system on your own boat if you want. Thus far no one has said they would make time to develop part of the system.

Obviously there are a lot of unknowns, but in principle, do you have time and would you enjoy taking on part of the system? This would be construction, writing code, liaising with authorities (if in scope).

For my own part, I want a solution in a year not five years. I want something I can use. A demonstrator too yes but usable.

All of you will have aspects, like me, that your are less interested in. So my question to you is; can you imagine a form of the project you would be prepared to put some time into? What bit appeals to you? If you only want to follow from sidelines and maybe interact on the forum with helpful suggestions, that is fine, but I need to know if I have any hands on tangible doers that have time and motivation to start this.

The style of project I would like to see would be stripped out to the minimum that will work. So indeed this may use standard drone components to just manually drive the dingy unmanned. It depends on who we have and what they like to do and what they can do.

It’s far better to have a little chat here about this now and work out that people don’t have time. This would be a positive result for me. I would know I’m on my own doing it. What would be a negative is leave the brain storming phase with promises of time which people are unable to deliver.

So probably you know now whether you will be able to find time for this. There is no shame in saying you don’t have time. You can still follow us or me as things go forward. But if in doubt please under promise to avoid a false start.

So who can identify an aspect of this project that they want and could make a tangible contribution to in principle?

 

[ridgy]

PM sent.

 

[sarabande]

The control/navigation aspects of an unmanned dinghy are solvable by piggybacking on current autonomous vehicle and vessel functionality .

Sadly, I can see immense legal constraints inhibiting the design phase, and as for the approval/insurance/compliance aspects...

I'm out, but best of luck. Don't invest too much money in the project.

 

[Baggywrinkle]

I work in automotive as a system architect ... many years of experience in embedded safety critical systems and also involvement in autonomous vehicles.

The first project I got excited about was this, called Promethius, this particular car was a collaboration between Jaguar and Lucas, the guy in the video is a good friend of mine ... I was writing embedded software for engine controllers at the time.

The date of the video was 1994 ... so coming up for 30 years ago.

As of today we are at the point where a car can just about drive safely along a motorway but that needs an alert driver as backup, A-roads are a challenge and depending on who you talk to they are either opposed due to possible legal consequences or going for it anyway to see what happens (Tesla). It won't happen until statistically self-driving cars are involved in fewer accidents than humans.

I am rapidly approaching retirement, and my take on your dinghy project is that an algorithmic approach, basically defining rules for the tender to follow and implementing them in SW will not work - they've been trying that for 3 decades in cars without the progress they had hoped for. The current approach is applying AI, gathering massive amounts of live driving data and applying AI to the problem. Maybe that might be the breakthrough required? Who knows.

The only devices I know capable of getting form A-B on their own without hitting anything are the current crop of drones, DJI probably being the most capable - and they get stuck and hit things quite often due to the constraints of their sensor systems, often the operator needs to take over to un-stick the drone.

Good luck with your project, it is possible IMO, but I'm not sure we currently have the computing power, knowledge, or sensor technology for a fully autonomous system in a crowded anchorage filled with swimmers, buoys, other small boats, fishing lines, submerged ropes, weeds, etc.

I would fullfill your requirements by taking the same approach as a drone, remote control and navigate the dinghy back to the mothership in the same way as a drone pilot would - some parts of the journey could be autonomous dependent on conditions, but a visual link and remote control to get it out of trouble is currently a must IMO.

 

[Zing]

I’ve designed CNC and robotic automation machinery with vision systems. If I was 30 yrs younger I’d love to help to do something like this, but essentially it’s an industry you are wanting to develop, not just a system. It’s a very ambitious project and it is feasible as all the tech is there to enable it. Good luck and I hope you do it.

Meanwhile, there is a low cost work-around for most situations. Use an anchoring arrangement to anchor just off the beach. I have one using two anchors and a block. One anchor on the beach, one with the block 20m off the beach and a double line to move the dink out of the surf. Very, very simple. Vulnerable to theft of course, which your plan improves on. Alternative solution, which I rarely use is a back up very lightweight dink with tiny engine to pull up onto the beach with chain and padlocks.

 

[Savannah]

Thanks for all your input. I am working with one other guy on this so be sure to understand that we will take a very pragmatic approach to try and get something working.

We have had a little brain storm and have concluded that we should avoid being too ambitious since many of you consider it too difficult. So my plan is to put together a ROV, initially without video back from the dinghy. In most cases I will be close to the boat - close enough to see. Bare in mind that the dinghy just needs to drop you at the nearest convenient point. You don’t need somewhere where the dinghy can be left.

Since I already have a lot of this problem solved for my big boat - I just need to make sure that a solution works for unmanned operation. I need longer range on the radio comms and there are a number of solutions for this. Ultimately video could be added over a gprs or 4g data connection but this is for later.

So accepting these limitations, makes everything more practical for a demonstrator and this will meet 90% of my needs.

The tricky bit in my opinion is the docking. A reliable system for docking the boat at the yacht end and a reliable means to wake up and undock the boat. This has to work every time without getting tangled up or stuck. Once the boat is docked we need to be sure it stays secured until next time we need it. It would be good to lift the dingy on a davit. So I have some ideas for this.

I will update progress on here. I don’t have any time table. The important thing is to enjoy the journey. If anyone else would like to help, please get I touch. Those of you that want to just follow from the sidelines, feel free to tag along.

C

 

[Rum Run]

These people are doing sea-drone type stuff on the Essex coast: Uncrewed Surface Vessel Design Build | SEA-KIT | Tollesbury
I have seen the vessel in person and it is as ugly as in the pictures!

 

[B27]

Do we actually want this in our harbours?

 

[Savannah]

Do we actually want this in our harbours?

Do we actually want this in our harbours?

Feel free to express your preference on that subject here.

 

[Baggywrinkle]

Most dinghies end up tied to the mothership by a piece of rope at anchor, perhaps an energise to release or motor driven twist to release pemanent magnet on the stern with an extendable painter on the dinghy with a suitable armature plate and holding/locating mechanism .... drive up, armature plate (attached to painter) aligns with the magnet and clamps on, then let out the painter and the dinghy drifts to a safe distance but remains firmly attached to the mothership via the magnet. To release the dinghy, trigger the release of the magnet and wind in the painter - dinghy free to leave.

 

[Daydream believer]

You do not need a magnet to attach. A pair of horizontal crab like claws, with a trigger in the centre, that bring the claws together, as the fixture (a large vertical ring perhaps, to allow a flexible coupling) on the bow touches it. Then all it needs is a servo, or lever, to over ride it & open the claws to release the dinghy. That will also need to hold the dinghy so personel can alight/dismount. So it would be mounted underneath a small aft deck so the dinghy bow docks below it & the crew drop down into the middle of the tender

Then as part of the development, start by using a radio transmitter, as with an RC plane & manually steering the dinghy to/from the mother ship.
Set up servos to control the motor. These would be start/stop. speed, direction etc. This will give data on what one needs when designing the robotic self drive system. There is no point designing software to control these functions until one actually knows how the manual parts will interact & what one wants them to do.
So set them up, then operate a transmitter from the shore & the mothership for a lot of hours first.
.
Some clients may be happy just being able to go ashore then manually sending the dinghy back using the transmitter, until they return. A system of multiple transmitters, so another crew member could then use the dinghy to go ashore somewhere else, & manually send it back to the mothership for the first crew to collect, would be better.

 

[Savannah]

You do not need a magnet to attach. A pair of horizontal crab like claws, with a trigger in the centre, that bring the claws together, as the fixture (a large vertical ring perhaps, to allow a flexible coupling) on the bow touches it. Then all it needs is a servo, or lever, to over ride it & open the claws to release the dinghy. That will also need to hold the dinghy so personel can alight/dismount. So it would be mounted underneath a small aft deck so the dinghy bow docks below it & the crew drop down into the middle of the tender

Then as part of the development, start by using a radio transmitter, as with an RC plane & manually steering the dinghy to/from the mother ship.
Set up servos to control the motor. These would be start/stop. speed, direction etc. This will give data on what one needs when designing the robotic self drive system. There is no point designing software to control these functions until one actually knows how the manual parts will interact & what one wants them to do.
So set them up, then operate a transmitter from the shore & the mothership for a lot of hours first.
.
Some clients may be happy just being able to go ashore then manually sending the dinghy back using the transmitter, until they return. A system of multiple transmitters, so another crew member could then use the dinghy to go ashore somewhere else, & manually send it back to the mothership for the first crew to collect, would be better.

Hello daydream believer. Would you like to take on any part of the project? Let me know if you wish to give sometime to it.