用自监督微调让机器人快速适应新环境，省去大量数据收集。
ELMP: Efficient Learning for Motion Planning via Analytical Policy Gradients
- 通过可微分运动学层直接优化策略，无需额外专家数据。
- 在未知环境中成功率从57.3%提升至89.8%，冷启动延迟降低近百倍。
- 适用于需要快速部署的机器人系统，尤其适合复杂机械臂场景。
神经运动规划器（NMPs）能实现快速反应式运动生成，但适应新环境通常需重新收集大量专家数据，计算成本高昂。本文提出ELMP框架，通过自监督微调实现数据高效适配。不依赖昂贵全局规划器生成新专家轨迹，而是利用可微分运动学层，直接以密集碰撞、目标到达和运动平滑性为目标优化策略。该方法将每样本适应成本降低约两个数量级。为增强对变化机械臂结构的泛化能力，引入点云显式编码工具几何信息。在经典与神经基线对比中，ELMP平均成功率达84.8%，冷启动延迟比传统方法低多个数量级。在未见环境中，自监督微调使成功率从零样本的57.3%提升至89.8%，彻底消除数据采集瓶颈。本方法保持毫秒级推理延迟，并在物理Franka Emika Panda机器人上验证有效。
原文摘要 · Abstract (English)
Neural Motion Planners (NMPs) enable fast reactive motion generation, but adapting them to new environments typically requires recollecting large expert datasets, which is computationally prohibitive. We propose ELMP, a framework for data-efficient adaptation via self-supervised fine-tuning. Rather than generating additional expert trajectories with expensive global planners, ELMP directly optimizes the policy through a differentiable kinematic layer using dense collision, target-reaching, and smoothness objectives. This replaces expert data generation with rapid problem sampling, reducing per-sample adaptation cost by roughly two orders of magnitude. To further support robust generalization across changing kinematic chains, we introduce a mechanism to explicitly encode tool geometry via point clouds. Benchmarked against classical and neural baselines, ELMP achieves an 84.8% average success rate with orders-of-magnitude lower cold-start latency than classical methods. In unseen environments, self-supervised fine-tuning improves success rate from 57.3% (zero-shot) to 89.8%, removing the data collection bottleneck. Our approach maintains millisecond-level inference latency and is validated on a physical Franka Emika Panda 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。