用在线规划实现复杂接触下的灵巧翻转，效果媲美强化学习。
DROP: Dexterous Reorientation via Online Planning
- 基于采样预测控制器与视觉位姿估计，实时搜索接触动作。
- 在真实机器人上实现与以往强化学习方法相当的翻转成功率。
- 适合对实时性要求高、需灵活应对接触场景的灵巧操作任务。
实现类人灵巧性是机器人领域长期挑战，尤其源于接触丰富系统在规划与控制上的复杂性。当前强化学习主流方法依赖大规模并行化、领域随机化的仿真环境，在海量接触条件下离线训练策略，以实现稳健的模拟到现实迁移。受近期实时并行仿真进展启发，本文探讨在线规划在接触丰富操作中的可行性，聚焦经典的手中立方体翻转任务。提出一种简洁架构：结合采样式预测控制器与基于视觉的位姿估计算法，在线搜索接触丰富的控制动作。通过全面实验评估方法在真实世界的表现、架构设计选择及鲁棒性关键因素，结果表明，该简单采样方法性能可媲美先前基于强化学习的工作。补充材料：https://caltech-amber.github.io/drop。
原文摘要 · Abstract (English)
Achieving human-like dexterity is a longstanding challenge in robotics, in part due to the complexity of planning and control for contact-rich systems. In reinforcement learning (RL), one popular approach has been to use massively-parallelized, domain-randomized simulations to learn a policy offline over a vast array of contact conditions, allowing robust sim-to-real transfer. Inspired by recent advances in real-time parallel simulation, this work considers instead the viability of online planning methods for contact-rich manipulation by studying the well-known in-hand cube reorientation task. We propose a simple architecture that employs a sampling-based predictive controller and vision-based pose estimator to search for contact-rich control actions online. We conduct thorough experiments to assess the real-world performance of our method, architectural design choices, and key factors for robustness, demonstrating that our simple sampling-based approach achieves performance comparable to prior RL-based works. Supplemental material: https://caltech-amber.github.io/drop.
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。