Skip to content

Dynamics and Perception Dataset of AutoDRIVE Ecosystem's Nigel Vehicle

License

Notifications You must be signed in to change notification settings

AutoDRIVE-Ecosystem/AutoDRIVE-Nigel-Dataset

 
 

Repository files navigation

AutoDRIVE Nigel Dataset

DOI

Dynamics and Perception Dataset of AutoDRIVE Ecosystem's 1:14 Scale "Nigel" Vehicle

⚠️ Dataset Size: This repository hosts the lite version of dataset (i.e. without camera frames), which is about 1.50 GB in size. The full version of this dataset is about 66 GB in size and is hosted seperately on Zenodo. Please check your internet data plan and local disk space before downloading the dataset.

Straight Track Skidpad Track
Fishhook Track Slalom Track
Eight Track Tiny Town Track

This repository uses AutoDRIVE Ecosystem to capture data from a 1:14 scale Ackerman-steered vehicle called Nigel. The source repository for AutoDRIVE Ecosystem can be found here.

Dataset Structure:

The vehicle dataset comprises the following:

DATA timestamp throttle steering leftTicks rightTicks posX posY posZ roll pitch yaw speed angX angY angZ accX accY accZ cam0 cam1 lidar
UNIT yyyy_MM_dd_HH_mm_ss_fff norm% rad count count m m m rad rad rad m/s rad/s rad/s rad/s m/s^2 m/s^2 m/s^2 img_path img_path array(float)

The traffic light dataset (only applicable for Eight Track) comprises the following:

DATA timestamp state
UNIT yyyy_MM_dd_HH_mm_ss_fff int{0=disabled,1=red,2=yellow,3=green}

Vehicle Parameters:

  • Wheelbase (m): 0.1415
  • Track width (m): 0.1530
  • Throttle Limit (norm%): 1.0000
  • Steering Limit (rad): 0.5236
  • Linear Velocity Limit (m/s): 0.2670
  • Angular Velocity Limit (rad/s): 0.8051
  • Throttle vs. Velocity Mapping:

Automated Data Collection:

The open_loop_control.py script makes use of AutoDRIVE Devkit's Python API. The script is capable of selecting a maneuver and its direction, and controlling the vehicle actuators within the prescribed limits in an open-loop setting.

python3 open_loop_control.py --maneuver={straight, skidpad, fishhook, slalom} --direction={cw, ccw} --throttle=[-1, 1] --steering=[0, 0.5236] --throttle_noise=[0, 0.001] --steering_noise=[0, 0.001]

Control Input Variations:

  • Throttle Gradations (norm%): 0.2, 0.4, 0.6, 0.8, 1.0 (straight maneuver has additional throttle gradations: 0.1, 0.3, 0.5, 0.7, 0.9)
  • Steering Gradations (rad): 0.1047, 0.2094, 0.3142, 0.4189, 0.5236 (straight maneuver does not have any steering gradations)

Data Visualization

Single Maneuver Data Visualization

Straight Maneuver Skidpad Maneuver
Fishhook Maneuver Slalom Maneuver
Eight Maneuver Parking Maneuver

Collective Maneuver Data Visualization

Straight Maneuver Skidpad Maneuver
Fishhook Maneuver Slalom Maneuver
Eight Maneuver All Maneuvers

Citation

We encourage you to read and cite the following papers if you use any part of this dataset for your research:

@article{AutoDRIVE-Ecosystem-2023,
author = {Samak, Tanmay and Samak, Chinmay and Kandhasamy, Sivanathan and Krovi, Venkat and Xie, Ming},
title = {AutoDRIVE: A Comprehensive, Flexible and Integrated Digital Twin Ecosystem for Autonomous Driving Research & Education},
journal = {Robotics},
volume = {12},
year = {2023},
number = {3},
article-number = {77},
url = {https://www.mdpi.com/2218-6581/12/3/77},
issn = {2218-6581},
doi = {10.3390/robotics12030077}
}

This work has been published in MDPI Robotics. The open-access publication can be found on MDPI.

@inproceedings{AutoDRIVE-Simulator-2021,
author = {Samak, Tanmay Vilas and Samak, Chinmay Vilas and Xie, Ming},
title = {AutoDRIVE Simulator: A Simulator for Scaled Autonomous Vehicle Research and Education},
year = {2021},
isbn = {9781450390453},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3483845.3483846},
doi = {10.1145/3483845.3483846},
booktitle = {2021 2nd International Conference on Control, Robotics and Intelligent System},
pages = {1–5},
numpages = {5},
location = {Qingdao, China},
series = {CCRIS'21}
}

This work has been published in 2021 International Conference on Control, Robotics and Intelligent System (CCRIS). The publication can be found on ACM Digital Library.

Languages

  • Jupyter Notebook 99.1%
  • Python 0.9%