用于原位采集多模态触觉感知数据的手持设备
机器人学
2015-11-13 v2
摘要
多模态触觉感知有望使机器人通过快速区分任务相关对象来提升其在操作任务中的表现。针对该触觉感知问题的数据驱动方法展现出前景,但缺乏合适的训练数据。在这篇两页论文中,我们提出一种便携手持设备,用于从自然场景(如家庭)中的对象高效获取多模态触觉感知数据。该设备上的多模态触觉传感器集成了基于织物的力传感器、接触式麦克风、加速度计、温度传感器以及加热元件。我们简要介绍我们的方法、描述该设备,并通过使用我们在某人家中浴室里接触7个任务相关对象所采集的小数据集进行评估来论证可行性。
引用
@article{arxiv.1511.03152,
title = {A Handheld Device for the In Situ Acquisition of Multimodal Tactile Sensing Data},
author = {Joshua Wade and Tapomayukh Bhattacharjee and Charles C. Kemp},
journal= {arXiv preprint arXiv:1511.03152},
year = {2015}
}
备注
This short 2-page paper was accepted and presented as a poster (https://goo.gl/8TqDTK) on Sep. 28, 2015 in IROS 2015 Workshop on 'See and Touch: 1st Workshop on multimodal sensor-based robot control for HRI and soft manipulation" organized by A. Cherubini, Y. Mezouar, D. Navarro-Alarcon, M. Prats, and J. A. Corrales Ramon. It was peer reviewed by 2 reviewers