Learning the References of Online Model Predictive Control for Urban Self-Driving

Author: Yubin Wang

Requirements

• Python 3.7.1
• PyTorch
• Wandb
• CasADi 3.5.5
• Unreal Engine 4.26
• Carla 0.9.12

Train your models

python main.py 

Evaluate the performance

Replace the model index with the model you trained to evaluate the performance:

python main.py --eval True --Loadmodel True --ModelIdex 35000

Citations

Please consider citing our paper if useful :

@article{wang2024learning,
  title={Learning the References of Online Model Predictive Control for Urban Self-Driving},
  author={Wang, Yubin and Peng, Zengqi and Xie, Yusen and Li, Yulin and Ghazzai, Hakim and Ma, Jun},
  journal={arXiv preprint arXiv:2308.15808},
  year={2024}
}

@article{wang2024chance,
  title={Chance-Aware Lane Change with High-Level Model Predictive Control Through Curriculum Reinforcement Learning},
  author={Wang, Yubin and Li, Yulin and Peng, Zengqi and Ghazzai, Hakim and Ma, Jun},
  journal={2024 IEEE International Conference on Robotics and Automation (ICRA)},
  year={2024}
}

References

|High-MPC| |Gym-CARLA| |SAC-Continuous-Pytorch |