CA-AC-MPC: CUDA-Accelerated Actor-Critic Model Predictive Control
For researchers and practitioners in robotics and control, this work addresses a computational bottleneck in differentiable MPC for RL, enabling faster training and deployment without sacrificing performance.
The paper introduces a CUDA-accelerated variant of actor-critic model predictive control (AC-MPC) that reduces training and inference latency while preserving control performance, achieving state-of-the-art lap times in agile drone racing.
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.