I don't believe that the Netduino has the processing power that you want to stream video over the network. I believe that you could get any wifi webcam to work with the robot, though. You would just need a little extra space in your power budget to account for the camera's additional power usage.
Depending on how soon you need the camera, I would just watch the winter sales on any of the big technology suppliers (Newegg, Tigerdirect, Amazon) for a deal on a wifi webcam that meets the specifications that you are looking for.
If you are looking to make your project entirely self-contained, then you may want to check out the CMUcam. It provides a good vision sensor, without having to do much processing on your Netduino chip. It can do most basic vision processing on-board, without the need for your chip to handle the processor intensive image processing.
I spent 13 years designing electronics of this exact nature: three phase induction motor reduced voltage soft starters and variable frequency AC drives. I spent the last few of those years as a VFD applications engineer helping customers select and configure this equipment for various loads and industries around the country as well.
You will not be able to build something that is cheap and safe. The voltages and currents involved are well beyond the safety margin of a hobbyiest, especially someone who is openly avoiding buying commercial units in order to save money. Don't do it!
While the theory behind AC motor control is very straightforward, the detail level work (heat sink sizing, snubbering, gate drive requirements, de-sat protection, motor overload calculations, bus capacitor protection, etc.) can be quite tricky to get down, especially with heavy duty cycling and regenerative power modes which a carnival ride will CERTAINLY be generating. I strongly caution you against trying to build something of this nature unless you have significant experience not only in microcontrollers and embedded systems design but also significant experience in power electronics and three phase circuitry. People get hurt and killed building this stuff.
My first question for you is whether speed control is really required, or if you only require a soft start up and slow down. Do you vary the speed of the motor once it is started? If not, you may be able to get away with a MUCH cheaper reduced voltage soft starter. These units act like three phase light dimmers; they only adjust the applied voltage to the motor. You will not have a lot of torque at low speeds, but with the right design of motor (NEMA class D) you can achieve exactly what you're after with a fraction of the cost and maintenance.
If you really do need to vary the full-load speed of the motor then you are more or less stuck using a variable frequency drive. As you are aware these are expensive and if you buy cheap you are likely to replace them sooner due to your high surge current (they call this "constant torque") application. What I would definitely recommend doing if this is the case would be to contact various manufacturers (Allen-Bradley, Cutler-Hammer, SAF drives, Benshaw, Yaskawa, etc.) and ask for reconditioned units. Ask for a drive capable of delivering 150% rated current for 30s (this is usually known as heavy duty) or size the drive 30-50% larger than your nominal current rating. You will also likely be running off of generator power which is notorious for being undersized and prone to brownouts and surges as the load requirements change with the state of the equipment being run. Drives don't like that (voltage sags cause current spikes as the motor starts slipping and surges can cause you to overvoltage the bus capacitors) and have a tendency to either fault out or blow up.
I am all about the little guy building something and saving a buck, but this is not the type of project to do this on. If you really want to build a three phase AC drive, start with a little 10HP 480V motor with a hand brake on a test bench. You have all the potential for experiencing the pants-filling sensation of an H-bridge failure or a bus capacitor explosion two feet from your head but without the potential lawsuits and loss of life (except perhaps your own).
Best Answer
One answer: Closed loop systems and/or sensor fusions
You can't just send the right commands to the motor and expect that the robot will act exactly as expected. Think about a very very small difference in the torque produced by your two motors that received two identical PWM signal (due to motor manufacturing tolerances). One will go slightly faster than the other and the robot will run along a circle (very big, but still a circle) instead of a straight line. The answer to this problem is closed loop control. Your motors, or your wheels will have encoders. And your control system will receive the signals from the encoders and will react in a way that ensure that your wheels are doing exactly what you expect them to do.
But in the case of an autonomous robot, this in not enough. Think about having a wheel that is slightly bigger than the other, because of wearing, or just due to manufacturing tolerances. Or think about a wheel that slips a little bit on the floor. In this case you don't know anymore for sure where you are. This method is called odometry or dead reckoning. In order to solve this, you have to use sensors that adds knowledge about your environment to your system. (Vision, proximity sensor, bumpers, magnetometer, gyroscopes, etc). Then, advanced algorithms are used to use the best of every sensor at each time step.