FAME-QP: A GPU-Accelerated Quadratic-Programming Solver for Real-Time Model Predictive Control
Overview
This paper introduces FAME-QP, a GPU-accelerated solver designed for dense quadratic programs in linear model predictive control. By combining constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs, the system automatically selects execution configurations via device calibration and problem-specific timing. It utilizes a cached operator bank to select initial penalties and updates operators lazily when problem matrices change. Numerical benchmarks and simulated control experiments on humanoid and quadrotor systems demonstrate reduced latency compared to competing GPU solvers while maintaining solution accuracy and control performance.
Generated from attributed reports · Updated 4 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
No current evidence-backed claims. Missing information remains not reported.
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalFAME-QP: A GPU-Accelerated Quadratic-Programming Solver for Real-Time Model Predictive Control
This paper introduces FAME-QP, a GPU-accelerated solver designed for dense quadratic programs in linear model predictive control. By combining constraint-dependent operator representations with fused single- and two-kernel iterations replayed through CUDA graphs, the system automatically selects execution configurations via device calibration and problem-specific timing. It utilizes a cached operator bank to select initial penalties and updates operators lazily when problem matrices change. Numerical benchmarks and simulated control experiments on humanoid and quadrotor systems demonstrate reduced latency compared to competing GPU solvers while maintaining solution accuracy and control performance.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).