I’m building an ocean going autonomous solar powered boat that has a Cube Orange running Rover 4.0 as its autopilot. I would like to integrate in an Iridium Short Burst Data (SBD) modem and service from Rock Seven so I can receive updates in the order of 2-4 times a day. In the updates I’d like to get the boats current position and heading plus status of a few things like batteries etc.
The Rock Seven RockBlock system looks perfect for the job but as I’m pretty new to ArduPilot I can’t figure out how to integrate the two. I have searched these forums and elsewhere and the only discussion/detail is the SPL solution as discussed here Global Satellite Telemetry for ArduPilot and here Stretching Comm Links: From Indoors to the Globe
Does anyone have any experience or anything to suggest on the subject? Is the SPL solution the only option, other than hacking something new together myself?
Should I physically wire the RockBlock’s UARt to the OrangeCube TELEM2? I am comfortable with UART<>UART wiring, its more about what happens next?
If so, what config do I need to do in Mission Planner?
I am comfortable with Arduino boards and wonder if I should be putting a SAMD21 based Arduino in between the CubeOrange TELEM2 port and the RockBlock…and have it filter/translate/reformat the messages etc?
How do I limit ArduRover to only send the few parameters I want and only the few times of day I want? Or do I do that with above mentioned Arduino?
How do I issue commands (like new waypoints) from Mission Planner to the boat?
Lots to think about, but I was hoping some of the obvious things that others have already learned could be shared? I’m happy to share what I learn as I go.
Sounds like we have embarked on a similar project, but you are significantly further down the process than I am. I was hoping you might have a minute to share your ideas.
How are you getting on? Would you mind sharing your thoughts and tips?
My current ideas are:
torpedo style for strength and stability
pi navigation and sensors
satellite comms with Lora backup
I am still questioning my choices of propulsion, either 2 thrusters or one with a rudder. I’m a bit worried about a rudder getting snagged or snapping off in a wave. I was really interested to see your post about the thrusters you got. Are you still pleased with them?
Thanks for getting in touch, I’m always happy to talk about this stuff with fellow builders😁
Do you mind if I ask where you are roughly located and where you are thinking of sending your boat? I’m in Sydney, Australia.
There is so much to consider with a project like this and I am sure I have over complicated my boat, but I am having a really good time learning about things and slowly building it up.
I have been making a ton of notes but haven’t blogged it anywhere, but I should in order to share back what I have learned, particularly since lots of it is built on the work of various others before me. I have also done a ton of work on HW, SW and electronics that I’ve developed myself. That said there are still so many design decisions to make!
In fact you have inspired me to have a go at writing it up online, so I will start that tonight and let you know where I put it.
For now though, here is a quick summary.
I’ve been working on this project since last June and thinking about it for some time before that. My current boat is a prototype, ideally I’d like to build a bigger one in future. It’s a 1.5 yacht style hull I built from some plans on the net.
I’m using a Cube Orange autopilot running ArduPilot Rover. It is absolute overkill (I know now), I could have gone with something much simpler and cheaper.
I have SIK radios for point to point telemetry from the Cube on the boat boat to my Windoze laptop that is running Mission Planner. I usually use MacOS, but Mission Planner only runs on Windows unfortunately and it seemed to be the best match (as opposed to QGround Control) for use with ArduPilot.
I also have a traditional RC TX/RX to manually control the boat during testing.
I am using LiPoFe4 batteries, 1 x 25Ah and 1 x 12Ah so far, but lots more work to do on that part of the project. I went SLA battery first but they are just too heavy and cannot provide the high currents or the deep cycle I think I need.
I have just purchased two Mauch power sensors and BECs to replace the ones that come from the Cube vendor (CubePilot) are not very good for low current measurement, quite disappointing as I bought two and they are not cheap.
I am running a Raspberry Pi 3B with the awesome Pi-Connect Lite Hat and Rpanion software from Steven at Rpanion.com. I’d highly recommend it. The Pi has a Huawei E3372 Cellular modem USB stick plugged into it. The Pi is also connected to the TELEM2 port on the Cube. Add in ZeroTier networking over the top and I have a working MAVLink over IP over Cellular connection between the Cube on the boat and my Mission Planner PC. Plan is to use that when the boat is beyond the range of the SIK radios.
Then the final comms path is via Rock7 and Iridium satellites. I’m still integrating this stuff, but getting close to having a 3rd comms method that will come into play when the boat goes over the horizon and out to sea.
So much more to talk about, but I will start a blog now, I have been meaning too for ages.
Feel free to ask questions though and I’m really keen to hear about what you are doing and how you get on over time.
Well done getting the blog started too. That will be an excellent source of inspiration and technical info. I’ll be sure to bookmark that!
I’m in Scotland which kinda makes it hard to launch a small boat as there are islands between me and the Atlantic so it is more likely to run aground. I think I’ll have to throw it in the water on the East coast and go up and round the top.
You are way ahead of me as mine is currently a thought experiment! It is also a hobby so I’m trying to do this on as small a budget as possible.
I have broken the project up into several areas and I’m taking copious notes for each in Evernote.
As I said, my thoughts so far are:
Torpedo style hull for strength and stability. I thought long and hard about the hull shape. It needs to be strong and buoyant, and low to the sea so it doesn’t catch the wind too much, so I thought of semi-submersible. I thought this team came up with a good idea, so I was going to try a variation of this. I discovered how expensive a boat hull is to buy, so I thought a large piece if PVC pipe is going to be much cheaper and stronger too. Having talked to folk on RC boat forums, salt water and moving parts don’t mix and they have told me at length about how short a time any electrical motor will last in the sea. This will need some thought!
Solar recharging to replenish the batteries. I’ve never had much joy with solar power or electrical things, so this needs a lot of work!
I just assumed I’d use a Pi for on-board brains as I’ve used one before. I like the Grove sensors as it means I can add lots of different sensors quickly and easily. I have some worries about navigation and coast avoidance. I would clearly have to set waypoints for the vessel to track a path around the world, but I’m concerned if it drifts off course either in a current or when the batteries run out and it is charging. The concern is that the boat drifts towards shore or wakes up in a location where there is land between it’s current location and the next waypoint. It would have to have an navigation override which would prioritise avoiding land rather than getting to the next waypoint. I did post this enquiry regarding using Dijkstra’s path planning but as yet there is no reply.
I had originally thought I could use The Things Network Community LoraWAN gateways as a way of getting data back from the boat on it’s travels. It would be free to connect to a gateway and send small packets. I didn’t really want to go down the expensive satellite communications path. Having posted in their form, it looks like this was slightly optimistic. I had over estimated the range of Lora and the amount of data which can be sent as part of their fair use policy. I wanted to cache data when there was no connection then upload when it found a gateway. It did spark an interesting discussion about using the motion sensor in the boat to sense when it is at the crest of a wave to ensure it has the best chance to get a connection though. I may still add Lora as a backup since it is cheap and very low power.
This is it so far. Nothing physical, all just thoughts, ideas and ambition!
I can’t wait to see some progress photos of your build on the blog site!