We're looking to build something like the LR Wave Glider - a surfboard and a ROV attached to it via a flexible tether 2m below it. The surfboard lets us have a waterline of 0 and putting all electronics below the surfboard means we avoid the constant bashing from the surf and waves on the surface.
The "sub" below will have a rudder and a T100 thruster for course correction and navigation.
The craft's max speed is about 1 knot, and maybe 3 knots in sea state 10/hurricanes.
We ran some numbers on the required rudder torque and for our initial testing 6Kgm would get us started.
The T100 motor is not used for constant navigation, however we expect to adjust the rudder from 60 times a minute to a few times an hour depending on the traffic. There will be a Raspberry Pi 3 connected but we would like the ability to turn it on/off remotely to conserve power.
Is this setup doable:
Arduino with a 3G/GPS shield connected to an Ardupilot APM, these 2 units are always on and responsible for navigation and course correction.
Can the RPi3 interface with and use the 3G and GPS shield on the Arduino Uno?
Can I power on/off the RPi3 via the Uno?
We will use Xbee for RC and eventually add satcom but currently the area in which we're testing has enough 2G/3G coverage out to sea and we'd like to take advantage of it.
Robin2:
An RPi should be able to communicate with the Arduino over a Serial connection.
I suggest you use a Mega simply because it has multiple HardwareSerial ports.
The Arduino should be able to control power to the RPi - perhaps via a relay or a transistor.
What size is your proposed boat/submarine?
...R
Great, a Mega is no issue, the APM is a modified Mega too.
The upper craft is a 9-10 feet surfboard, the sub below with the motor/rudder will be 6-8 feet in length.
jackrae:
Sea state "10" you say. Sea state 9 (the highest listed in wikipedia) is waves in excess of 46 feet (14 metres)
How do you propose to sail a craft in those conditions
6KgM (43 foot-pounds) seems a very high torque for a small rudder on a vessel doing a nominal 1knot
it appears to be a combination of height of the sea and frequency.
rapid 1 foot swells would seem to do more than one 30 foot swell in the same period
especially if the tether is only 2 meters.
I think the idea is pretty neat. wonder what it would do if encountered any of those lost ducks....What Can 28,000 Rubber Duckies Lost at Sea Teach Us About Our Oceans?
I can see that propulsion for free is great, but slow. a wind driven boat on nice days could to 5 knots....
once you get this working, the platform just screams for modification.
jackrae:
How do you propose to sail a craft in those conditions
6KgM (43 foot-pounds) seems a very high torque for a small rudder on a vessel doing a nominal 1knot
dave-in-nj, the RPi is so we can run an RTL-SDR to receive AIS broadcasts for navigation and anything else that needs CPU for post processing, having it on board opens up data collection/processing possibilities.
The fact that we now have free propulsion allows us to greatly lower power requirements and we don't need speed, just remote and self navigation ability. I'm a network engineer and a familiar OS (linux) lets me get work done faster so if I can turn it off when not needed I can keep my power requirements low. I live next to a coastline that receives 4.5 - 5 KWh/m2/day and we just want the craft to loiter in a predefined area. Using the APM allows us to use the rich ecosystem created around Ardupilot for ground control systems and things like geo-fencing.
jackrae, there is no intention of navigating the craft in to 14m swells, it was merely to give an idea of how slow it is. The craft has no sail, in fact it has nothing on top and this is exactly what allows us to survive in the ocean. However, the LR wave glider has traveled through hurricanes collecting data and transmitted it via Iridium all while riding 14m waves so it's definitely within the realm of possibilities for this craft. It manages because the sub below behaves like an anchor and prevents the entire arrangement from being yanked clear of the water in to the air.
The rudder torque was a back of the hand calulation, well Excel sheets with the formula already in there and we could probably go a lot lower for our first prototype as long as we keep weight down.
Do you have a way to keep both processors from accessing the RTC at same time? Then yes. Easier with I2C, with SPI will need to add control signal buffering so only one processor drives the signals at a time.
Otherwise no, the control signals will interfere with each other.
