Data collection tools with active data acquisition algorithm implemented for CARLA Simulator.
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.
-
Ubuntu 20.04
-
Python 3.8+
-
CARLA 0.9.15
-
CUDA 11.7
-
Download Carla version 0.9.15 here.
-
Build conda environment
cd CARLA_ADA conda create -n carla38 python=3.8 & conda activate carla38 pip3 install -r requirements.txt
-
Start carla, at carla root directory
./CarlaUE4.sh
-
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
. -
Make data (image, lidar) to a dataset with labels
python format_helper.py -s <raw_data_directory>
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 |
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.
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}
}
-
Code and Rules
-
Experiments, and parameters setting
-
Paper work