Parameter Estimation using Hybrid Gradient Descent

    •  Johnson, R.S., Di Cairano, S., Sanfelice, R., "Parameter Estimation using Hybrid Gradient Descent", IEEE Conference on Decision and Control (CDC), DOI: 10.1109/​CDC45484.2021.9682794, December 2021.
      BibTeX TR2021-146 PDF
      • @inproceedings{Johnson2021dec,
      • author = {Johnson, Ryan S. and Di Cairano, Stefano and Sanfelice, Ricardo},
      • title = {Parameter Estimation using Hybrid Gradient Descent},
      • booktitle = {IEEE Conference on Decision and Control (CDC)},
      • year = 2021,
      • month = dec,
      • doi = {10.1109/CDC45484.2021.9682794},
      • url = {}
      • }
  • MERL Contact:
  • Research Areas:

    Control, Dynamical Systems


In this paper, we consider the problem of estimating a vector of unknown constant parameters for a hybrid system whose flow and jump dynamics are affine in the unknown parameter. Using a hybrid systems framework, a hybrid algorithm is proposed and sufficient conditions are established to guarantee exponential convergence of the parameter estimate. Robustness of the proposed algorithm with respect to measurements noise is analyzed, and examples are provided showing the merits of the proposed approach.