Subsea Intervention ROV control

Hello everyone
I have spent numerous hours on the net searching for answers to my Arduino questions.
I am sure there are som very bright brains here that will know how to guide me on my way.

Short story:
I am designing a remotely Operated Vehicle ROV. The ROV will have a tether to the surface where it will be controlled via a laptop and QGC Control software. The signals are converted from ethernet to HomePlug at the surface and converted back again inside the ROV. The signals go via an unshielded twisted pair cable.
I do have one twisted pair in spare.
Operating the ROV with motors lights and camera is all fine.

The challenge:
What I want to do next, set beside the ROV operating system, is to use Arduino to control extra servos, lights, and motors. The servos and motors and lights will all be subsea connected to the ROV and I want to be able to set power them on and off, then control the speed and rotation of the motors and position of the servos, dim the lights, etc from the surface.
To control them I was hoping to use a simple control board with pots and buttons or even a PS4 controller.

can someone aloborate on this ?

Thank yoy

To control the the motors, servos and light switches

Where is the Arduino, and what should it do?

How much of the project is ready, "working on the bench"?

HomePlug is a mains voltage device.
Are you putting 110 vac on one of the cable pairs?

At either end use an Arduino with an ethernet adaptor and connect via a switch/hub to the HomePlug. It'll share the link with the QGC Control software.

Pass commands from one Arduino to the other using HTTP or something like that.

What is a “HomePlug”?

"To control them I was hoping to use a simple control board with pots and buttons or even a PS4 controller."

Use the form search function in the upper right of this page and search for key terms like "ROV servo" to find years of past similar projects.

There is also the Open Rov project, are you aware of it? It's a good source of information. Unless you have good reason and the capability to re-invent the wheel, following some tried-and true methods/examples should be more productive.

If as you say,

Operating the ROV with motors lights and camera is all fine.

, why don't you expand the system you already seem to have/conceived/control of?

Thank you all for your replies. I am aware of the open ROV project, yes. The ROV control is working. not only on the bench but in the water as well. This new ROV is still on the drawing board and I do not wnt to temper with the ROV QGC stable version. I want the control system for extra motor, light and servos to be a seperate system (It`s for tools and add ons, not for ROV control)

I was hoping that someone could give an specific way of approach on how-to best start. A, B, C ....

If you can only use a single twisted pair for communication then the first step is to evaluate the performance of that link while the cable is under water and the adjacent pairs are simultaneously carrying data.

Put a 1 kHz square wave into the sending end using a line driver and have a look on an oscilloscope to see what it looks like at the receiving end. If it looks bad you could try adding termination resistors and a low pass filter.

Alternatively use a current loop driver (4-20mA) to send data or use analog levels to communicate.

"I do have one twisted pair in spare."

I assume that is two wires. If so, use one wire for a common return ground and the other as a command tx line to the rx in the ROV.

That makes sense, and having a working system already and being aware of other open source projects means your way ahead, brilliant.

What mikb55 said, plus I would add as a starting point, think about a data rate you would like to have that is suitable for control and hopefully allows for reasonable future expansion, the type of hardware/signal type/method (as mikb55 said) & communications method/protocol that seems most suitable (eg simplex, half-duplex, full duplex, and various messaging schemes/protocols), and suits the budget/any time constraints, based on testing as suggested.

Fortunately (or unfortunately depending how you look at it), there are a wide range of ways of getting data up and down a single twisted pair, ranging from rolling your own at component level with suitable devices, to commercial boxes or embeddable modules for RS485 or RS422 line driver/receiver/transparent links, transparent links/extenders for ethernet/lan or usb, to modem-type devices, or even just single ended with line driver/receiver as suggested.

Then there's the choice of messaging scheme, ranging from writing your own at low level, to using perhaps similar or same/transparently/compatible with QGC / drone /mimic protocol (if it can be), or some other "wrapper" scheme.

As an example, many moons ago I worked on a variety of land-based conmmercial ROV's, and one type used a single twisted pair in the tether for comms with RS485 hardware for half-duplex, 9-bit serial (can't remember bit/baud rate, but it might have been around 38400), with a low-level industrial messaging scheme, and was perfectly adeqate for motion control as well as all the extras such as lighting, pan-tilt, camera commands etc. and telemetry/acks in the other direction. The video was analog via 75ohm coax and we couldn't see any interference.