Hi Jacky,
Thank you for your help so far! My comments below.
Do not use the files I posted on the board, since it still uses the old TFDU4100. I will upload my current board version with the changes metioned below sometime next week. It should be relatively easy to forward port them to your board. I also wrote several test programs that I will add to the upload.
I understand, but the board that I posted is based on a version of your board that uses TFD4300 and not TFDU4100. Myself and some other teams also already ordered that board and the parts. You can see the files
here. So I really hope we can fix the board not having to make a new one...

1.) The BR/D line (Pin2) on the TOIM4232 was not working correctly when controlled from DTR of the FT232. I wrote a test program to drive DTR and it changed etc., but the TOIM4232 did neither receive or transmit. After several hours of checking things out I still could not find the problem. It all looked fine with the scope on the pins.
I changed the BD/R line and connected it to GND instead of DTR and it works fine now. I can transmit and receive with the circuit. But without the ability to drive BR/D, we can not reprogram the baudrate and are limited to 9600 baud. I will try a pull down next Monday.
I see. The BR/D (pin 2) of TOIM4231 connects to DTR (also pin 2) of FT232RL through a via in the board that I posted. So it seems easy to use a wire and add a pull down or even to connect to GND and cut the connection to DTR if this is the case. We could also try to calculate if baud rate 9600 is such a big problem or not, because this is already much faster than last year anyway.
2.) My concern is still the angle of the IRDA transceiver since it is limited to 30 deg. and 1m length as per the IRDA standard. However, I think we may be able to add our own IR diode and drive it. If I have time I will investigate this further.
This is also easy to add to the current project as well.
3.) I am wondering whether I should connect RTS to RESET on the TOIM4232. This is assuming that I get 1.) to work correctly. I found out that it is impossible to read the current board rate, so it may be better to be able to reset the TOIM4232 to a known baud rate
Well, for RESET removing the USB cable and plugging it back again does not sound as such a big deal for me either.
I am currently very busy and probably will not be able to do much more work on this. I will try and send you my working board so that you can test it with the actual robots to determine distance and angle for the reception.
We will be glad to test. Our boards and components should be arriving soon as well, so maybe if you give me the updated circuit schematic I can try to fix the board. I want to make a video and pictures showing to the teams how to build the board once we succeed figuring out a working design.