用扩散模型生成更符合物理规律的控制轨迹，提升离线决策可靠性。
Model-Based Diffusion Sampling for Predictive Control in Offline Decision Making
- 将规划器与动力学模型交替更新，逐步修正轨迹可行性。
- 在D4RL和DSRL基准上均超越已有扩散方法，真实机器人验证有效。
- 适合需要高可靠轨迹的机器人控制场景，尤其适用于数据稀缺任务。
通过扩散模型进行离线决策时常生成与系统动力学不符的轨迹，限制了其在控制中的可靠性。我们提出模型预测扩散器（MPDiffuser），一种组合式扩散框架，将扩散规划器与动力学扩散模型结合，生成任务对齐且动力学合理的轨迹。MPDiffuser在采样过程中交替更新规划器与动力学模型，逐步修正轨迹可行性同时保留任务意图。一个轻量级排序模块筛选最满足任务目标的轨迹。该组合设计提升了样本效率与适应性，使动力学模型可独立利用多样且未见过的数据。实证表明，MPDiffuser在无约束（D4RL）和有约束（DSRL）基准上持续优于先前扩散方法，并在真实四足机器人上成功部署验证。
原文摘要 · Abstract (English)
Offline decision-making via diffusion models often produces trajectories that are misaligned with system dynamics, limiting their reliability for control. We propose Model Predictive Diffuser (MPDiffuser), a compositional diffusion framework that combines a diffusion planner with a dynamics diffusion model to generate task-aligned and dynamically plausible trajectories. MPDiffuser interleaves planner and dynamics updates during sampling, progressively correcting feasibility while preserving task intent. A lightweight ranking module then selects trajectories that best satisfy task objectives. The compositional design improves sample efficiency and adaptability by enabling the dynamics model to leverage diverse and previously unseen data independently of the planner. Empirically, we demonstrate consistent improvements over prior diffusion-based methods on unconstrained (D4RL) and constrained (DSRL) benchmarks, and validate practicality through deployment on a real quadrupedal robot.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。