Real-time robotic control demands fast action generation. However, existing generative policies based on diffusion and flow matching require multi-step sampling, fundamentally limiting deployment in time-critical scenarios. We propose Dispersive MeanFlow Policy Optimization (DMPO), a unified framework that enables true one-step generation through three key components: MeanFlow for mathematically-derived single-step inference without knowledge distillation, dispersive regularization to prevent representation collapse, and reinforcement learning (RL) fine-tuning to surpass expert demonstrations. Experiments across RoboMimic manipulation and OpenAI Gym locomotion benchmarks demonstrate competitive or superior performance compared to multi-step baselines. With our lightweight model architecture and the three key algorithmic components working in synergy, DMPO exceeds real-time control requirements (>120Hz) with 5-20x inference speedup, reaching hundreds of Hertz on high-performance GPUs. Physical deployment on a Franka-Emika-Panda robot validates real-world applicability.

翻译：实时机器人控制需要快速的动作生成。然而，现有的基于扩散和流匹配的生成策略需要多步采样，这从根本上限制了其在时间关键场景中的部署。我们提出了弥散均值流策略优化（DMPO），这是一个统一的框架，通过三个关键组件实现真正的一步生成：MeanFlow用于无需知识蒸馏、通过数学推导的单步推理，弥散正则化以防止表示崩溃，以及强化学习（RL）微调以超越专家演示。在RoboMimic操作任务和OpenAI Gym运动基准测试中的实验表明，与多步基线方法相比，DMPO具有竞争力或更优的性能。凭借我们轻量级的模型架构以及三个关键算法组件的协同工作，DMPO以5-20倍的推理加速超过了实时控制要求（>120Hz），在高性能GPU上达到数百赫兹。在Franka-Emika-Panda机器人上的物理部署验证了其实际应用价值。