多视角融合提升机器人抓取效率，同时保持对单视角故障的鲁棒性。
Merging and Disentangling Views in Visual Reinforcement Learning for Robotic Manipulation
- 通过融合多视角信息并引入单视角特征，实现高效策略学习
- 在Meta-World和ManiSkill3上提升样本效率，部署更轻量
- 适合需要高鲁棒性和低资源消耗的机器人视觉控制场景
视觉在机器人操作中广泛用于视觉伺服。由于世界具有三维特性，使用多个相机视角并融合它们能生成更好的表示，从而提升Q-learning性能，训练出更高效的策略。然而，这些多视角策略对摄像头失效敏感，且部署负担重。为此，我们提出一种融合与解耦算法（MAD），能高效融合视角以提高样本效率，同时通过在多视角特征输入中加入单视角特征来解耦视角，提升策略鲁棒性并支持轻量级部署。我们在Meta-World和ManiSkill3上验证了该方法的高效性与鲁棒性。
原文摘要 · Abstract (English)
Vision is well-known for its use in manipulation, especially using visual servoing. Due to the 3D nature of the world, using multiple camera views and merging them creates better representations for Q-learning and in turn, trains more sample efficient policies. Nevertheless, these multi-view policies are sensitive to failing cameras and can be burdensome to deploy. To mitigate these issues, we introduce a Merge And Disentanglement (MAD) algorithm that efficiently merges views to increase sample efficiency while simultaneously disentangling views by augmenting multi-view feature inputs with single-view features. This produces robust policies and allows lightweight deployment. We demonstrate the efficiency and robustness of our approach using Meta-World and ManiSkill3. For project website and code, see https://aalmuzairee.github.io/mad
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。