Active Data Acquisition in Autonomous Driving Simulation

Data collection tools with active data acquisition algorithm implemented for CARLA Simulator.

News

2024.1.10 Fix label tool and format helper, adapter for pcdet.

2023.12.20 Finish L2 and L4.

2023.9.16 Finish L1 and L3 part, at ./label_tools/lidar_tool.

2023.6.23 Finish the main framework design, and basic framework of algorithm.

Dependency

• Ubuntu 20.04

• Python 3.8+

• CARLA 0.9.15

• CUDA 11.7

Installation

1. Download Carla version 0.9.15 here.

2. Build conda environment

   cd CARLA_ADA
   conda create -n carla38 python=3.8 & conda activate carla38
   
   pip3 install -r requirements.txt

Usage

Data Recording

1. Start carla, at carla root directory

   ./CarlaUE4.sh

2. Start recorder, execute the command in the root directory

   python3 data_recorder.py

   The configuration can be changed by modifying the files in the folder config.

3. Make data (image, lidar) to a dataset with labels

   python format_helper.py -s <raw_data_directory>

Data and Result

Dataset Name	Total Frames	Map
D1	900	Town02
D1-S	579	Town02
D2	900	Town02
V	375	Town03

Dataset Name	Vehicle	Training Time	mAP
D1	996	14min40s	0.508
D1-S	996	11min32s	0.492
D2	1835	16min07s	0.647

Contributing

Thank you for your interest in contributing to this project! Contributions are highly appreciated and help improve the project for everyone. If you have any questions or need further assistance, please feel free to open an issue.

Acknowledgments

• CARLA Simulator
• CARLA Ros Bridge
• CARLA_INVS

Citation

The paper lags behind the code

@article{Lai2023ActiveDA,
 title={Active Data Acquisition in Autonomous Driving Simulation},
 author={Jianyu Lai and Zexuan Jia and Boao Li},
 journal={ArXiv},
 year={2023},
 volume={abs/2306.13923}
}

Todo List

• Code and Rules

• Experiments, and parameters setting

• Paper work