基于多尺度扩散的可扩展规划
Extendable Planning via Multiscale Diffusion
AI总结：
针对现有扩散规划器受训练轨迹长度限制难以处理长时域规划的问题，本文提出两阶段多尺度扩散方案，结合渐进式轨迹扩展、分层多尺度Diffuser及单模型统一架构，实现了性能更优的长时域高效决策。
摘要：
长时域规划在复杂环境中至关重要，但Diffuser等基于扩散的规划器受限于训练过程中观测到的轨迹长度，这造成了一个两难困境：有效规划需要长轨迹，但长轨迹会降低模型性能。本文提出了这一可扩展长时域规划挑战，并给出两阶段解决方案：第一，渐进式轨迹扩展通过多轮组合拼接增量构建更长的轨迹；第二，分层多尺度Diffuser通过跨时间尺度推理，实现长时域上的高效训练与推理。为避免需要多个独立模型，我们提出自适应规划思考与递归HM-Diffuser，将分层规划统一到单个模型中。实验表明，我们的方法取得了显著性能提升，推动了长时域下可扩展、高效决策的发展。
英文摘要：
Long-horizon planning is crucial in complex environments, but diffusion-based planners like Diffuser are limited by the trajectory lengths observed during training. This creates a dilemma: long trajectories are needed for effective planning, yet they degrade model performance. In this paper, we introduce this extendable long-horizon planning challenge and propose a two-phase solution. First, Progressive Trajectory Extension incrementally constructs longer trajectories through multi-round compositional stitching. Second, the Hierarchical Multiscale Diffuser enables efficient training and inference over long horizons by reasoning across temporal scales. To avoid the need for multiple separate models, we propose Adaptive Plan Pondering and the Recursive HM-Diffuser, which unify hierarchical planning within a single model. Experiments show our approach yields strong performance gains, advancing scalable and efficient decision-making over long-horizons.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
