msma's Introduction

MSMA

we focus on traffic scenarios where a connected and autonomous vehicle (CAV) serves as the central agent, utilizing both sensors and communication technologies to perceive its surrounding traffics consisting of autonomous vehicles, connected vehicles, and human-driven vehicles.

Overview

Gettting Started

1. Clone this repository:

git clone https://github.com/xichennn/MSMA.git
cd MSMA

2. Create a conda environment and install the dependencies:

conda create -n MSMA python=3.8
conda activate MSMA
conda install pytorch==1.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge

# install other dependencies
pip install pytorch-lightning
pip install torch-scatter torch-geometric -f https://pytorch-geometric.com/whl/torch-2.1.0+cu121.html

3. Download the CARLA simulation data and move it to the carla_data dir.

Training

In train.py, There are 3 hyperparameters that control the data processing:

mpr: determines the mpr of the connected vehicles in the dataset
delay_frame: determines the latency ranging from 1 to 15 frames (0.1~1.5s)
noise_var: determines the Gaussian noise variance ranging from 0 to 0.5 \

and there are two in the model arguments that control the data fusion:

commu_only: when set to true, only data from connected vehicles are utilized
sensor_only: when set to true, only data from AV sensors are utilized
when both commu_only and sensor_only are set to False, data from both sources will be integrated

Results

Quantitative Results

Metrics	MPR=0	MPR=0.2	MPR=0.4	MPR=0.6	MPR=0.8
ADE	0.62	0.61	0.59	0.59	0.56
FDE	1.48	1.47	1.40	1.37	1.33
MR	0.23	0.22	0.22	0.21	0.20

Qualitative Results

MPR=0	MPR=0.4	MPR=0.8

Citation

If you found this repository useful, please cite as:

@article{chen2024msma,
  title={MSMA: Multi-agent Trajectory Prediction in Connected and Autonomous Vehicle Environment with Multi-source Data Integration},
  author={Chen, Xi and Bhadani, Rahul and Sun, Zhanbo and Head, Larry},
  journal={arXiv preprint arXiv:2407.21310},
  year={2024}
}

License

This repository is licensed under Apache 2.0.

msma's People

Contributors

Watchers

msma's Issues

Error on Running the `train.py`

The detailed error goes as follows:

(msma) user@user-Precision-3660:~/Documents/1/MSMA$ python train.py 
Traceback (most recent call last):
  File "train.py", line 25, in <module>
    train_set = scene_processed_dataset(root,
TypeError: __init__() got an unexpected keyword argument 'delay_frame'

Recommend Projects