ConfVE

Introduction

ConfVE is the first configuration testing approach in the ADS domain, which serves as a testing framework that utilizes scenarios from pre-existing ADS scenario-generation techniques and a genetic algorithm to produce alternative configurations to identify emerged failures in an ADS by preventing the masking of failures and maximizing the possibility of producing bug-revealing violations.

The video recordings of ConfVE when different types of violations happen are available at

Hardware and Software Requirements

• Intel Core i9 12900K (16-core)

• 96 GB memory and above

• Ubuntu 22.04 and above

• Docker-CE version 19.03 and above

• Python 3.11 and above

• NVIDIA RTX 3090 and above (Optional)

• NVIDIA driver version 455.32.00 and above (Optional)

• NVIDIA Container Toolkit (Optional)

Additional Information

ConfVE runs multiple Docker container simultaneously, therefore the requirement for running ConfVE varies based on the number of containers (default=5) you wish to run.

Installing

In this section we will be discussing steps to replicate the results discussed in the paper

|--DIR_ROOT
    |--apollo
        |--modules
        |...
    |--ConfVE
        |--data
        |--src
            |--config.py
            |--main.py

1. Git clone this repository or download the zip file of this project. Unzip the ConfVE.zip file to get the data and src directories.

2. Place the /ConfVE under the same directory with /apollo

INSTALLING Apollo ADS

1. Git clone or Download the Apollo 7.0 from https://github.com/yuntianyi-chen/apollo-baidu/tree/master, and name the apollo root directory /apollo

2. Copy the directory ConfVE/data/module/sim_control to apollo/modules/

3. Replace the file apollo/docker/scripts/dev_start.sh with the file of ConfVE/data/scripts/apollo_multi_container/dev_start.sh,

4. At the root directory of Apollo, start up Apollo container via bash docker/scripts/dev_start.sh

5. Enter the docker container using bash docker/scripts/dev_into.sh

6. In the container, build Apollo via ./apollo.sh build or ./apollo.sh build_opt_gpu

INSTALLING ConfVE

1. Enter the ConfVE folder, install the required Python libraries via pip install -r requirements.txt. This includes installing the required Python libraries: numpy, Shapely, pandas, networkx, docker, and cyber_record.

2. Customize the parameters in ConfVE/src/config.py according to your requirements, or just remain default.

3. Place your directories of initial record files under ConfVE/data/records, run python3 prepare.py

The folder name should follow the rule of ApproachName_MapName(e.g., DoppelTest_borregas_ave). We've provided a group of sample records file under ConfVE/data/records for your reference.

4. Execute sudo chmod -R 777 apollo/data to change the permission of the directory apollo/data

5. Run python3 main.py to start the testing

After running ConfVE for extended period of time, you should see temporary record files of scenarios generated under apollo/data/records. This is also the step to replicate the results presented in the paper.

Notice

• For the first run, the map parser would be automated executed to generate and save the map info file. If you test on a large map, it may take a long time.