Python and MATLAB codebase for performing second-order system identification from time-domain data of a systems dynamic response to a decreasing step input function.

Physically-informed model discovery of systems with nonlinear, rational terms using the SINDy-PI method. Contains functionality for spectral filtering/differentiation.

Sparse variational Gaussian process state-space models in PyTorch & Pyro. Research-grade, ready-to-use GP-SSM baseline.

Explore the double-descent phenomena in the context of system identification. Companion code to the paper (https://arxiv.org/abs/2012.06341):

For my study of the use of local measurement data in multi-vehicle formation control policies.

For research into the application of Koopman operators at Boston University.

Optimization of Laguerre-Volterra networks for nonlinear dynamic system identification using diverse metaheuristics

This project is a system information app. It can indentify CPU, Private IP address and hardware platform, number of threads and hostname.

An experimental code base for system identification with Jax.

Deep learning scripts used to accurately predict human-robot system dynamics for future prosthesis control methods

For my personal study of simple real/complex Fourier methods and their application to various systems.

Python and MATLAB codebase for obtaining a magnitude and phase frequency response, and a corresponding transfer function estimate, from measured voltage data of two sensors using a NI DAQ/ LabVIEW experimental setup.

Online active identification algorithm for linear dynamical systems.

Adaptive oscillators can be described as a mathematical tool able to synchronize with a rhythmic and periodic signal by continuously estimating its fundamental features (i.e. frequency, amplitude, phase, and offset).

PID Tunining with deep neural network from scratch

Rex is a JAX-powered framework for sim-to-real robotics.

Jiles-Atherton system identification tool: Given a B(H) curve, finds the Jiles-Atherton model coefficients. Supports various JA model formulations. Has an interactive GUI.

This App reads dynamic responses and fit the parameters of a lumped-mass dynamic structural system. The structure is formed by known masses conected by unknown springs, dampers and inerters.

Transfer operator toolbox for numpy/torch.

A SOFA environment is created for a tripedal soft legged robot model with Python scripts to be used for RL training along with system identification using PSO and GA algorithms for material parameters optimization.