arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2607.02845cs.ROcs.AIcs.CV
收藏

受差分放大器启发的AmpAttention用于多视图机器人操作

Differential Amplifier-Inspired AmpAttention for Multi-View Robotic Manipulation

National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, and Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University(西安交通大学人机混合增强智能技术国家级重点实验室、人工智能与机器人研究所)

Jin Yang, Ping Wei, Nanning Zheng

AI总结:

针对机器人视图图像问题,提出受模拟电路差分放大器启发的AmpAttention机制,抑制注意力噪声。基于此引入RVAF模型,提升训练效率与任务性能,还扩展为RVAF++,在高精度任务上成果显著。

摘要:

具有注意力机制的多视图机器人操作方法在训练效率和任务性能方面取得了显著进展。然而,机器人视图图像中固有的冗余、遮挡和视点依赖性常常导致严重的注意力漂移。为了应对这一挑战,我们提出了AmpAttention,一种受模拟电路中的差分放大器启发的新型注意力机制。它旨在抑制注意力噪声并捕获高信噪比信号以实现更可靠的感知。基于此,我们引入了RVAF模型,该模型集成了任务引导的视图内和视图间AmpAttention。与先前的最先进方法相比,RVAF在18个RLBench任务(249个变体)中实现了最佳平均成功率,同时将训练时间减少了33.3%。RVAF在现实世界的高精度任务中也显示出强大的潜力,例如它能够拿起飞镖并准确地将其插入红色靶心。此外,我们通过合并SAM2图像编码器将RVAF扩展到RVAF++。RVAF++在高精度任务上取得了显著进展,在“插入钉子”任务上实现了91%的成功率。更多定性结果可在匿名项目网站this https URL上获得。

英文摘要:

Multi-view robotic manipulation methods with the attention mechanism have recently achieved significant progress in both training efficiency and task performance. However, the inherent redundancy, occlusion, and viewpoint dependency in robotic view images often lead to severe attention drift. To address this challenge, we propose AmpAttention, a novel attention mechanism inspired by differential amplifiers in analog circuits. It aims to suppress attention noise and capture high signal-to-noise ratio signals for more reliable perception. Based on this, we introduce the RVAF model, which integrates task-guided intra-view and inter-view AmpAttention. Compared to previous state-of-the-art methods, RVAF achieves the optimal average success rate across 18 RLBench tasks (249 variations) while reducing training time by 33.3\%. RVAF also demonstrates strong potential in real-world high-precision tasks, exemplified by its ability to pick up a dart and accurately insert it into the red bullseye. Furthermore, we extend RVAF to RVAF++ by incorporating the SAM2 image encoder. RVAF++ achieves substantial gains on high-precision tasks, achieving a 91\% success rate on the `insert peg' task. More qualitative results are provided at the anonymous project website https://anonymous.4open.science/w/RVAF-Anonymization.

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。

↑