v1.0.0

NeuralMPCX v1.0.0 — First Stable Release

First stable release of NeuralMPCX, a flexible Model Predictive Control (MPC) framework with neural network integration.

Highlights

• Neural MPC with LSTM-based dynamic models and CasADi integration
• Classic Linear MPC with state-space models
• Classic Nonlinear MPC (NMPC) with symbolic CasADi dynamics
• Constraint handling (state, input, terminal, soft constraints)
• Warm-starting and multi-start support
• Augmented Kalman Filter for state and bias estimation
• MIMO transfer function to state-space conversion utilities

Examples

• Cascaded Two-Tank System (neural MPC)
• Grinding Circuit (linear MPC)
• CSTR (nonlinear MPC and neural MPC)

Installation

pip install git+https://github.com/hzdr/neural-mpcx.git