cs.RO · 2605.29155 Copy arXiv ID · May 27, 2026 Save CA-AC-MPC: CUDA-Accelerated Actor-Critic Model Predictive Control Authors: Antoonio Buo , Vittorio Cammarota , Michele Avagnale , Pierluigi Arpenti , Vincenzo Lippiello , Fabio Ruggiero
Abstract In the literature, actor-critic model predictive control (AC-MPC) integrates MPC with reinforcement learning to enable high-performance control of complex dynamical systems. However, its differentiable MPC layer requires repeatedly solving an optimization problem in both the forward and backward passes, leading to substantial training and inference latency. This paper tackles this bottleneck introducing a CUDA-accelerated variant that significantly reduces end-to-end execution time while preserving the control performance of the baseline formulation. Simulation results on an agile drone racing task show that our approach achieves state-of-the-art lap times and near-limit dynamic behaviour with markedly reduced training and inference time.
Explore similar work Jun 23, 2026 · Gabriel Bravo-Palacios, Jianghan Zhang, Zachary Pestrikov +2 Model Predictive Control Combustion Control
May 6, 2026 · Jonathan Spieler, Sven Behnke Model Predictive Control Dream
Aug 4, 2026 · cs.RO J/K move · Enter open · S save
Babak Akbari, Melissa Greeff
Robora Lab, Department of Electrical and Computer Engineering Queen's University, Kingston, ON, Canada
Model Predictive Control (MPC) delivers constraint-aware control, but its reliance on online optimization limits its use on systems with fast dynamics, high-dimensional models, or long horizons. Existing GPU implementations typically treat the device as a linear-algebra accelerator, leaving the optimization loop dependent on repeated kernel launches and high-latency memory transfers. This paper introduces CUDA MPC, a GPU-native MPC framework that co-designs the optimization algorithm, execution model, and memory architecture for CUDA hardware. CUDA MPC pairs a parallel-in-horizon alternating direction method of multipliers (ADMM) splitting with a fused CUDA kernel that runs the entire iterative solve on the device. Intermediate optimization variables stay in low-latency, on-chip shared memory, and a localized atomic-flag protocol synchronizes only adjacent horizon blocks, minimizing host intervention, kernel-dispatch overhead, and global-memory traffic. Across six nonlinear robotics benchmarks spanning increasing state dimension and constraint density, CUDA MPC sustains real-time rates at horizons one to two orders of magnitude longer than CPU solvers: it solves an optimization-based collision-avoidance parking problem with 100 s of lookahead within a 0.1 s sampling interval, and is the only solver evaluated that achieves both real-time execution and collision-free coordination for a centralized 10-agent swarm, where acados and CasADi return no feasible solution and require 3.5 s and 4.5 s per solve. Against tensor-framework implementations of the same ADMM splitting, the fused kernel is up to
965 × 965\times 965 × faster.