Autonomous sonar equiped boat help choosing hardwa

Forum 2005-2010 (read only) Hardware Interfacing
1

I'm not sure if I posted in the right section. But here it goes.

I'm a mechanical engineering student and I have a project to build for my mechatronics class. My team and I chose a pretty big project to integrate with the school's robotics club. The project is an autonomous boat that would make topographic maps of the bottom of the river/lake/pool.

Right now we're thinking of basing the thing on a sonar and gps. Before venturing out, the water area would be cut into manageable squares with preset parallel paths through those squares. That could be done on a computer before setting it free. Once on the water it would follow the parallel lines and take depth readings with a sonar and coupling those with the gps coordinates am and the yaw of the boat to then saving it. The boat would then run for a while on it's self before coming back. You could then take the raw data and process it on a computer and could then make a topographic map!

We all know how to program in C, I have some experience with the arduino platform, I did a few projects and one of the guys has some experience with sonar, so it should be too too bad.

Sooooo, I would like some help to size up how much computing power is needed for something like this. It would need, a pin for temperature, 2 for sonar, yaw, compass, 2 motor drivers, 1 servo, gps and writing to SD. On top of that, the program will probably be quite long so I guess a mega board would be wise. I found adafruit's gps/logging shield Adafruit GPS logger shield kit [v1.1] : ID 98 : $19.50 : Adafruit Industries, Unique & fun DIY electronics and kits witch would simplify a lot of things. Anyone know if it works with the mega ?

Anyhow, I would like to hear what some people with more experience think of this project and if it would be implementable with that hardware? Would having a netbook running the arduino as an I/O be overkill?

2

Anyone know if it works with the mega ?

Not without some hacking: the SD card interface uses SPI, which has a slightly different pinout on the Mega.

You should do some research on GPS precision: I think you'll find that the accuracy of a "consumer-grade" GPS is not good enough.

3
  1. I love this project - have been brain-grinding for a couple of years before getting onto Arduino. To survey a dam for water volume you need to a) make a loop around the dam at minimum navigable depth, then b) spiral in on a parallel course e.g. 5 m from the previous lap. I'm working on DGPS on NMEA streams at the moment - I think 2 or 3 m accuracy is fine for a typical water body. The depth sounder is an issue though - really need 10 - 20 cm accuracy with varying bottoms: hard, muddy, weedy.
  2. there are a lot of implementation issues to work out. I haven't seen a library that abstracts steering so you don't have to re-code between left-right power (tank or catamaran) and vectored steering (rudder). How to manage 1000 or so trackpoints in memory so you can pick up the loop - etc.

I hope you can progress - I'll post some data on NMEA DGPS soon.

4

Hi
I have actually made a little boat to map the water depths in lakes here in Norway. You can read about my project and maybe have a look at my Arduino sketches here.
http://letsmakerobots.com/node/21098