Skip to content

Latest commit

 

History

History
21 lines (18 loc) · 1.43 KB

README.md

File metadata and controls

21 lines (18 loc) · 1.43 KB
Raw

NMPC-DCLF-DCBF-CDC2021

In this paper, we discuss the feasibility, safety and complexity about existing optimal control laws using control barrier functions. This is the reference implementation of our paper.

If you find this code useful in your work, please consider citing:

@inproceedings{zeng2021nmpc-dclf-dcbf,
  title={Enhancing Feasibility and Safety of Nonlinear Model Predictive Control with Discrete-Time Control Barrier Functions},
  author={Zeng, Jun and Li, Zhongyu and Sreenath, Koushil},
  booktitle={2021 IEEE Conference on Decision and Control (CDC)},
  year={2021}
}

Instructions

Several test files are created to reproduce the results in the paper:

  • testFeasibilityMPCDCBF.m compares the feasibility performance between MPC-DCBF and NMPC-DCBF with different hyperparameters.
  • testFeasibilityMPCGCBF.m compares the feasibility performance between MPC-GCBF and NMPC-DCBF with different hyperparameters.
  • testFeasibilityDCLFDCBF.m compares the feasibility performance between DCLF-DCBF and NMPC-DCLF-DCBF.
  • testSafety.m compares the safety performance between MPC-CBF, MPC-GCBF and NMPC-DCBF.

Notice that running either testFeasibilityMPCDCBF.m or testFeasibilityMPCGCBF.m or testFeasibilityDCLFDCBF.m will take more than 4 hours due to the weak memory management using matlab together with excessive Yalmip calling. To check results or have a quick visualization, feel free to load data directly from the /data.