Design Unknowns
The unknowns in this project are fairly evident. First and foremost, a suitable internet connection needs to be established, and the hardware for this connection needs to be broken down to a point where an FPGA can interface with it. This could prove to be more difficult than it sounds due to the rather limited options for wireless internet connections in the Seattle area. However, we are still researching this issue, and hope to have a solution (or viable work-around) soon.
Also there is the issue of the bandwidth of the connection. Some wireless internet connections operate faster than T1, where others can only reach data rates about two times faster than dial-up. The requirements for our connection depend largely on the resolution of the camera we use, the speed that it operates, and the amount of compression we can implement on the signal. The communication in the other direction (the user inputs) will not contribute greatly to the bandwidth requirements for this project, but one could easily see how this might not be true for more sophisticated vehicles. We want to find a connection that offers at the very least a 1Mbps data transfer rate.
The only other unknown that I can think of is the web based interface, and how exactly that will work to decode the image data from the vehicle, and send user input (probably keyboard strokes) to the car at the same time. Perhaps a javascript applet running 2 threads? This will have to be researched and discussed too.