fault-tolerant-control

PyTorch + Gymnasium framework for aerospace control. Deep RL (SAC, PPO, DDPG, DQN, IHDP, ET-DHP) and adaptive controllers (AIDI, MPC, iADP) on nonlinear F-16/B747/UAV/satellite sims. Includes in-flight damage modeling, real-time 3D flight viewer, TensorBoard/WandB, Optuna. EN/RU docs.

Adaptive structural-survivability and SHM research platform for aerospace systems: damage localization, remaining-control-authority analysis, bounded multi-action recovery, resource/uncertainty constraints, executed-command-consistent held-out verification, deterministic replay, HIL campaigns, and a P1 physical-validation path.

Fault-tolerant supervisory control for autonomous UAV swarms - CTDE-MAPPO, ROS2, PyBullet, IEEE paper under review 2026

Open-access Simulink implementation and back-EMF lookup-table data for the method proposed in the IEEE TTE paper on open-phase-tolerant online current references for maximum torque range and minimum loss, with current and torque-ripple limits, for n-phase nonsalient PMSMs with nonsinusoidal back EMF (DOI: 10.1109/TTE.2023.3288525).

MATLAB numerical simulation for self-adjusting prescribed-performance control of saturated quadrotor UAVs.

AI-Based Fault Diagnosis and Fault-Tolerant Control of Permanent Magnet Synchronous Motor (PMSM) for Electric Vehicles

Open-access Simulink model for per-phase stator-resistance estimation by DC injection in fault-tolerant six-phase induction-motor drives (IEEE TIE, DOI 10.1109/TIE.2024.3417992).

EE650 (Basics of Modern Control Systems) course project: fault-tolerant control of a quadrotor after complete rotor loss. Residual Koopman (EDMDc) MPC on the Mueller-D'Andrea relaxed-hover model, and certified (LMI) fault-detection-delay budgets. Python simulator validated against published results; simulation only.

Resilient fractional-order sliding-mode control of a redundant rail-mounted SCARA robot under actuator faults and joint locking: Python/MATLAB code, raw results and RoboDK digital twin.

Simulation research on neural aircraft recovery, concurrent failures, bounded runtime assurance, and Z3 verification.

Residual-Gated Advantage IHDP Reinforcement Learning technique for online adaptive flight control under faults (code + paper)

A Cyber-Physical Sovereign Architecture for Autonomous Grid-Inspection Humanoids: Formal Verification, Lyapunov Dissipation Bounds, and Zero-Trust Sandboxing (IEC 63439 & IEC 62443 Compliant).