You've explored the possibilities of walking robots, flying robots, and sailing robots. The final frontier is robots that can actually maneuver under the water. It only makes sense to use the same techniques that you've mastered to explore the undersea world. In this section, I'll explain how to use the capabilities that you have already developed in a Remote Operated Vehicle (ROV) robot. There are, of course, some interesting challenges that come with this type of project, so get ready to get wet!

As with the other projects in this chapter, there are possibilities of either buying an assembled robot or assembling one yourself. If you'd like to buy an assembled ROV, try http://openrov.com. This project, funded by Kickstarter, provides a complete package, albeit with electronics based on BeagleBone Black. If you are looking to build your own robot, there are several websites that document possible instructions for you to follow, such as http://dzlsevilgeniuslair.blogspot.dk/search/label/ROV. Additionally, http://www.mbari.org/education/rov/ and http://www.engadget.com/2007/09/04/build-your-own-underwater-rov-for-250/ show platforms to which you can add Raspberry Pi.

Whether you have purchased a platform or designed your own, the first step is to engage Raspberry Pi to control the motors. Fortunately, you should have a good idea of how to do this, as Chapter 5, Creating Mobile Robots on Wheels, covers how to use a set of DC motor controllers to control DC motors. In this case, you will need to control three or four motors based on which kind of platform you build. Interestingly, the problem of control is quite similar to the quadcopter control problem. If you use four motors, the problem is almost exactly the same; except instead of focusing on up and down, you are focusing on moving the ROV forward.

There is one significant difference: the ROV is inherently more stable. In the quadcopter, your platform needed to hover in the air, a challenging control problem because the resistance of air is so small and the platform responds very quickly to changes. Because the system is so dynamic, a microprocessor is needed to respond to the real-time measurements and to individually control the four motors to achieve stable flight.

This is not the case underwater, where our platform does not want to move dramatically. In fact, it takes a good bit of power to make the platform move through the water. You, as the operator, can control the motors with enough precision to get the ROV moving in the direction you want.

Another difference is that wireless communication is not available to you underwater, so you'll be tethering your device and running controls from the surface to the ROV through wires. You'll need to send control signals and video so you can control the ROV in real time.

You already have all the tools at your disposal for this project. As noted from Chapter 5, Creating Mobile Robots on Wheels, you know how to hook up DC motor controllers; you'll need one for each motor on your platform. Chapter 4, Adding Vision to Raspberry Pi, shows you how to set up a webcam so that you can see what is around you. All of this can be controlled from a laptop at the surface, connected via a LAN cable and running vncserver.

Creating the basic ROV platform should open the possibility of exploring the undersea world. An ROV platform has some significant advantages. It is very difficult to lose (you have a cable attached), and because the device tends to move quite slowly, the chances for catastrophic collisions are significantly less than many other projects. The biggest problem, however, is keeping everything dry!