Skip to content

install.md

Latest commit

 

History

History
147 lines (94 loc) · 2.77 KB

install.md

File metadata and controls

147 lines (94 loc) · 2.77 KB

Installation Guide for ApolloSimFuzz

This document provides step-by-step instructions for installing and setting up the environment for ApolloSimFuzz.

✅ Pre-requisites

Before installation, please ensure the following dependencies are available on your system:

⚠️ GPU support is required for running Baidu Apollo modules.

🛠️ Installation

Step 1: Install Core Dependencies

We provide a unified bash script that automatically performs the following steps:

  • Downloads Baidu Apollo
  • Creates the corresponding Conda environment
  • Builds the Docker image for TrafficSandbox

Run the installation script:

bash install.sh

If the installation is successful, you should see the following output:

[ApolloSimFuzz] Installation completed successfully.
[ApolloSimFuzz] Can use conda environment: apollosimfuzz-7.0

After installation, the following artifacts should be available:

  • A new directory:

    apollo/

  • A new Docker image (verify using docker images):

    drivora/sandbox:latest

🧱 Build Baidu Apollo

Apollo must be compiled manually after the initial setup.

Step 1: Navigate to the Apollo directory

cd apollo

Step 2: Apply required patches

Apply the patches located in:

apollo_bridge/patches/

Specifically:

  • Move apollo_bridge/patches/dev_start_ctn.sh to:

    apollo/docker/scripts/

  • Replace the following file to resolve build issues:

    apollo/WORKSPACE

    with:

    apollo_bridge/patches/WORKSPACE

Step 3: Start the Apollo development container

bash docker/scripts/dev_start.sh

You should see a success message indicating that the container has started correctly.

Step 4: Enter the container and build Apollo

bash docker/scripts/dev_into.sh
./apollo.sh build

⏱️ The build process may take a significant amount of time depending on your hardware configuration.

🚗 Build TrafficSandbox (Optional)

If the TrafficSandbox Docker image was not successfully built during the initial installation, you can build it manually as follows.

Step 1: Enter the TrafficSandbox directory

cd TrafficSandbox

Step 2: Build the TrafficSandbox Docker image

bash build.sh

After the build completes, verify the image:

docker images | grep drivora/sandbox

You should see the following output:

drivora/sandbox   latest

🧪 Test Demo

Check and modify config.yaml and then execute:

python start_fuzzer.py