Skip to content
/ MI-HPO Public

Model Parameter Identification via a Hyperparameter Optimization Scheme (MI-HPO)

License

Notifications You must be signed in to change notification settings

hynkis/MI-HPO

Repository files navigation

Model Parameter Identification via a Hyperparameter Optimization Scheme for Autonomous Racing Systems

Abstract

In this letter, we propose a model parameter identification method via a hyperparameter optimization scheme (MI-HPO). Our method adopts an efficient explore-exploit strategy to identify the parameters of dynamic models in a data-driven optimization manner. We utilize our method for model parameter identification of the AV-21, a full-scaled autonomous race vehicle. We then incorporate the optimized parameters for the design of model-based planning and control systems of our platform. In experiments, MI-HPO exhibits more than 13 times faster convergence than traditional parameter identification methods. Furthermore, the parametric models learned via MI-HPO demonstrate good fitness to the given datasets and show generalization ability in unseen dynamic scenarios. We further conduct extensive field tests to validate our model-based system, demonstrating stable obstacle avoidance and high-speed driving up to 217 km/h at the Indianapolis Motor Speedway and Las Vegas Motor Speedway. The source code for our work and videos of the tests are available at https://github.com/hynkis/MI-HPO.

Comments

Video demonstration

  • Field tests at the Indianapolis Motor Speedway (IMS) and Las Vegas Motor Speedway (LVMS) Youtube video

Usage

# Run model identification. A result would be saved as '.csv' in results directory.
python3 run_model_identification.py

# Plot the learned model with data.
python3 plot_learned_model.py

Results of Model Parameter Identification (Sample tire dataset)

Since the dataset of the race vehicle is confidential, we created a random (but reasonable) sample tire dataset to run our codes.

References

Please consider to cite this paper in your publications if this repo helps your research: https://arxiv.org/abs/2301.01470

@article{seong2023model,
  title={Model Parameter Identification via a Hyperparameter Optimization Scheme for Autonomous Racing Systems},
  author={Seong, Hyunki and Chung, Chanyoung and Shim, David Hyunchul},
  journal={IEEE Control Systems Letters},
  year={2023},
  publisher={IEEE}
}

We refer the following for Hyperband:

Disclaimer

For any question, please contact Hyunki Seong.

About

Model Parameter Identification via a Hyperparameter Optimization Scheme (MI-HPO)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages