FAME-QP: A GPU-Accelerated Quadratic-Programming Solver for Real-Time Model Predictive Control

Researchers presented FAME-QP, a GPU-accelerated solver for the dense quadratic programs that arise in linear model predictive control, combining constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs. FAME-QP automatically selects its execution configuration, according to the presentation. The work targets real-time model predictive control, where solver speed on dense quadratic programs is the binding constraint.

We present FAME-QP, a GPU-accelerated solver for dense quadratic programs arising in linear model predictive control. FAME-QP combines constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs, automatically selecting the execution config