Lyapunov-stable Neural Control for State and Output Feedback

In this project, an observer in the form of a stable neural network is proposed for any nonlinear MIMO system. As a result of experience, this observer utilizes a nonlinear in parameter neural network (NLPNN) which unlike LPNN, supports systems with higher degree of nonlinearity with no pre-knowledge of its dynamics. The learning rule for this n…

Runtime safety net for LLM agents. Detects token spirals, kills doomed tasks early, tells you exactly why. Rust core, Python SDK. pip install state-harness

A mobile robot equipped with a 3D camera moves in a room with four circular areas, and it has to localise and classify four objects which are positioned inside every area. A robotic arm with 6-DoF picks up the object and places it in a basket according to its class

Research on Control of Cyber-Physical Systems

Official Implementation of L4DC paper: Compositional Neural Certificates for Networked Dynamical Systems

The Project involves 3 typical convex optimization problems in control and the SDP (Semi-definite Programming) form or in other words the LMI (Linear Matrix Inequalities) form of each problem is achieved analytically and then the SDP optimization problems are solved in MATLAB using MATLAB CVX toolbox.

📕Code for paper Parallelizable Complex Neural Dynamics Models for PMSM Temperature Estimation with Hardware Acceleration.

Efficient Computation of Lyapunov Functions Using Deep Neural Networks for the Assessment of Stability in Controller Design

A collection of definitions, theorems, lemmas, results, rules, techniques, derivations, proofs and more from the field of control theory. The note management system "Obsidian" is used for note management.

AI-Driven Control Strategy for Differential Drive Wheeled Mobile Robots: Neural Network-Based Parameter Optimization and Real-Time Stabilization for Multi-Waypoint Navigation.

Real time parameter estimation on Grey Box Dynamic Systems

VNAE: A geometric approach to global stability in massive multi-agent systems with asymmetric dissipation, validated at scale.

LQR-LQG-Controller Design

Second Order State Hallucination (SOSH) mitigates attacks in formation control of multi-agent systems. Awarded the "People's Choice Award" at the 24th Annual Highschool Research Symposium.

Real-time volatility sensitive detection and correction for any sequential generative process. AutoTune, feedback learning, reflexive analysis, Monte Carlo SDE bands, Kalman filtering, GARCH variance modeling, signal detection, and domain anchoring.

Formalization of foundational Control Theory in Lean (early attempts)

3D Reconstruction of an indoor environment using monocular camera-equipped differential mobile robot: An oriented path-tracking approach integrating hybrid control strategies.

This repository contains the complete supplementary material for the canonical/quadratic example of the VNAE framework.