Receive UDP data on NVIDIA Jetson using matlab coder

I am having a problem with generating elf files running on NVIDIA JETSON using the UDP block.

The project I am using is using the following tools:

- MATLAB Coder Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms

- NVIDIA Jetson Orin Nano with JetPack 5.1.1 installed.

I am creating a project that receives data packets from UDP for processing, I use the UDP Receive block in the MATLAB Coder Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms library as shown

I did the simulation and the model worked fine. But when I created PIL to create elf file to run on NVIDIA Jetson Orin Nano, it gave me the following error

/usr/bin/ld: ../DAHostLib_Network.o: in function `CreateUDPInterface':

DAHostLib_Network.c:(.text+0x10): undefined reference to `CreateTargetUDPInterface'

/usr/bin/ld: ../DAHostLib_Network.o: in function `DestroyUDPInterface':

DAHostLib_Network.c:(.text+0x30): undefined reference to `DestroyTargetUDPInterface'

collect2: error: ld returned 1 exit status

make: *** [gpu_proc_x1.mk:320: gpu_proc_x1.elf] Error 1

I followed the error message to find the DAHostLib_Network.c file, this library file was created by matlab coder on NVIDIA Jetson for compilation.

- I tried to use TCP/IP transmission using the block as shown in the figure

- When using TCP/IP block, I successfully build an elf file to run on NVIDIA Jetson

The problem is that the data I receive is in UDP protocol, and currently the UDP block cannot build an elf file to run on NVIDIA Jetson.Please show me how to solve this problem so that I can receive UDP packets on NVIDIA Jetson

Answer 1

[Abhiram]

Hi @Thanh Tran Van,

I see that you are having issues when generating elf files running on NVIDIA JETSON while using the UDP Receive block.

In general, PIL mode is designed for algorithm verification on the target hardware, not for using hardware peripheral blocks or device driver blocks.

For more guidance on the configuration and use of PIL simulation, refer to the MATLAB Documentation page given below:

https://www.mathworks.com/help/ecoder/ug/configure-and-run-pil-simulation.html