Power cycling just the camera when initialization fails does not help.
We performed the documented diagnostics on the TTL serial port and the TTL to RS-232 converter board, via the loopback jumpers on both and the DDT, and they both passed.
I'm wondering if the serial port receive interrupt handlers aren't properly reinitialized when the RC's RESET button is pressed, but are reinitialized fine when the entire RC is power cycled, or the OI ROBOT RESET button is pressed.
I'm thinking this might be, because we built a 3-pin camera serial to DB-9 cable, and were able to view the commands being sent to the camera when initialization failed. We saw that the "CR" command for STATE #3 was indeed being sent by the RC properly. Either the camera wasn't ACKing it, or the RC wasn't picking up the ACK.
I THINK the camera was ACKing the "CR" commands, because when I connected to the camera directly with a laptop (immediately after initialization from the RC failed), and issued equivalent "CR" commands from the laptop, the camera ACKed them just fine.
Still just trying to find out if this is an isolated incident though. And right now, it looks like it is.
