Add Troubleshooting Guide for Simulink API

Author: Tinker-Twins

README.md

@@ -66,6 +66,9 @@ This section describes the Simulink API for AutoDRIVE, which can be used to deve
    - **[OPTIONAL]** To build the `jar` file yourself, it is recommended to use [Apache Maven](https://maven.apache.org/download.cgi) (tested with version 3.8.1) with [Java Development Kit](https://www.oracle.com/java/technologies/downloads/?er=221886#java8) (tested with version 8u411). Maven will automatically take care of downloading the [`Java-WebSocket`](https://github.com/TooTallNate/Java-WebSocket) library and neatly package everything into a single file (an "uber jar") based on the [`pom.xml`](autodrive_simulink_api/lib/pom.xml). Once the `mvn` command is on your path, simply `cd` to the `lib` directory and execute the `mvn package` command.
 2. Add the `autodrive_simulink_api` directory to MATLAB path by right-clicking on it from MATLAB's file explorer and selecting `Add to Path` → `Selected Folders and Subfolders`.
 
+> [!NOTE]
+> If you face issues installing/running the AutoDRIVE Simulink API or examples, check out the [Troubleshooting Guide](autodrive_simulink_api/troubleshooting/README.md) for helpful tips.
+
 ## USAGE
 
 Run the vehicle-specific `example` file:

autodrive_simulink_api/README.md

@@ -20,6 +20,9 @@ This directory hosts the Simulink API for AutoDRIVE, which can be used to develo
    - **[OPTIONAL]** To build the `jar` file yourself, it is recommended to use [Apache Maven](https://maven.apache.org/download.cgi) (tested with version 3.8.1) with [Java Development Kit](https://www.oracle.com/java/technologies/downloads/?er=221886#java8) (tested with version 8u411). Maven will automatically take care of downloading the [`Java-WebSocket`](https://github.com/TooTallNate/Java-WebSocket) library and neatly package everything into a single file (an "uber jar") based on the [`pom.xml`](lib/pom.xml). Once the `mvn` command is on your path, simply `cd` to the `lib` directory and execute the `mvn package` command.
 2. Add the `autodrive_simulink_api` directory to MATLAB path by right-clicking on it from MATLAB's file explorer and selecting `Add to Path` → `Selected Folders and Subfolders`.
 
+> [!NOTE]
+> If you face issues installing/running the AutoDRIVE Simulink API or examples, check out the [Troubleshooting Guide](troubleshooting/README.md) for helpful tips.
+
 ## USAGE
 
 Run the vehicle-specific `example` file:

autodrive_simulink_api/troubleshooting/Build Toolchain Setup.png

@@ -0,0 +1,46 @@
+# Troubleshooting Guide
+
+This directory contains troubleshooting tips for the AutoDRIVE Simulink API, which can help debug build/runtime errors, and assist in installing the necessary dependencies and setting up the correct configuration.
+
+## Potential Errors:
+
+Some of the examples require code generation, which may lead to build errors if not installed.
+
+![](No%20Compiler%20Error.png)
+
+Some of the examples require GPU code generation, which may lead to build errors if not installed properly.
+
+![](MEX%20Compiler%20Error.png)
+
+The MEX compiler setup is version-specific and can lead to version-mismatch errors if not installed properly.
+
+![](Visual%20Studio%20Version%20Error.png)
+
+## Potential Fixes:
+
+Download and install the [MATLAB Support for MinGW-w64 C/C++/Fortran Compiler](https://www.mathworks.com/matlabcentral/fileexchange/52848-matlab-support-for-mingw-w64-c-c-fortran-compiler) support package. This can also be done using MATLAB Add-On Explorer.
+
+Setup the GPU MEX compiler by following the steps below:
+
+1. Install Microsoft Visual Studio with `Desktop development with C++` workload selected [Tested with Visual Studio 2019 (V16.11.49)]
+
+    ![](Visual%20Studio%20Installation.png)
+
+2. Install NVIDIA CUDA [Tested with CUDA 11.8 (V11.8.89)]
+
+    > [!NOTE]
+    > It is important to install CUDA after installing Visual Studio so that CUDA can configure Visual Studio settings during installation.
+
+    ![](CUDA%20Installation.png)
+
+3. Setup the MEX Compiler by running `mex -setup C` and `mex -setup C++` in MATLAB Command Window and selecting `Microsoft Visual C++ 2019 (C)` as the compiler.
+
+    ![](MEX%20Compiler%20Setup.png)
+
+4. Set the Simulink build process toolchain to `NVIDIA CUDA (w/Microsoft Visual C++ 2019) | nmake` (selecting `Automatically locate an installed toolchain` should choose this already)
+
+    ![](Build%20Toolchain%20Setup.png)
+
+> [!WARNING]
+> - MATLAB/Simulink requires specific versions of Visual Studio and CUDA to be installed. Please refer to the official GPU Coder requirements [here](https://www.mathworks.com/support/requirements/gpu-coder.html).
+> - MATLAB/Simulink is known to have trouble working with Visual Studio 2022. Please refer to an official thread about this issue [here](https://www.mathworks.com/matlabcentral/answers/2158705-mexcuda-can-t-find-visual-studio-2022-but-mex-can?s_tid=srchtitle).