Suppr超能文献

如何通过可互操作的编程接口推动医疗保健领域的社交机器人用例。

How can social robot use cases in healthcare be pushed - with an interoperable programming interface.

机构信息

Institute for Medical Informatics BFH, Berne University of Applied Sciences, Höheweg 80, CH, 2502, Biel, Switzerland.

出版信息

BMC Med Inform Decis Mak. 2023 Jul 11;23(1):118. doi: 10.1186/s12911-023-02210-7.

Abstract

INTRODUCTION

Research into current robot middleware has revealed that most of them are either too complicated or outdated. These facts have motivated the development of a new middleware to meet the requirements of usability by non-experts. The proposed middleware is based on Android and is intended to be placed over existing robot SDKs and middleware. It runs on the android tablet of the Cruzr robot. Various toolings have been developed, such as a web component to control the robot via a webinterface, which facilitates its use.

METHODS

The middleware was developed using Android Java and runs on the Cruzr tablet as an app. It features a WebSocket server that interfaces with the robot and allows control via Python or other WebSocket-compatible languages. The speech interface utilizes Google Cloud Voice text-to-speech and speech-to-text services. The interface was implemented in Python, allowing for easy integration with existing robotics development workflows, and a web interface was developed for direct control of the robot via the web.

RESULTS

The new robot middleware was created and deployed on a Cruzr robot, relying on the WebSocket API and featuring a Python implementation. It supports various robot functions, such as text-to-speech, speech-to-text, navigation, displaying content and scanning bar codes. The system's architecture allows for porting the interface to other robots and platforms, showcasing its adaptability. It has been demonstrated that the middleware can be run on a Pepper robot, although not all functions have been implemented yet. The middleware was utilized to implement healthcare use cases and received good feedback.

CONCLUSION

Cloud and local speech services were discussed in regard to the middleware's needs, to run without having to change any code on other robots. An outlook on how the programming interface can further be simplified by using natural text to code generators has been/is given. For other researchers using the aforementioned platforms (Cruzr, Pepper), the new middleware can be utilized for testing human-robot interaction. It can be used in a teaching setting, as well as be adapted to other robots using the same interface and philosophy regarding simple methods.

摘要

简介

对当前机器人中间件的研究表明,大多数中间件要么过于复杂,要么已经过时。这些事实促使我们开发一种新的中间件来满足非专业人士的可用性需求。所提出的中间件基于 Android 构建,旨在置于现有的机器人 SDK 和中间件之上。它在 Cruzr 机器人的 Android 平板电脑上运行。已经开发了各种工具,例如通过 web 界面控制机器人的 web 组件,这使其使用更加便捷。

方法

中间件使用 Android Java 开发,并作为应用程序在 Cruzr 平板电脑上运行。它具有一个 WebSocket 服务器,与机器人接口,并允许通过 Python 或其他与 WebSocket 兼容的语言进行控制。语音接口利用 Google Cloud Voice 文本到语音和语音到文本服务。该接口使用 Python 实现,允许与现有的机器人开发工作流程轻松集成,并开发了一个 web 界面,通过 web 直接控制机器人。

结果

创建了新的机器人中间件,并将其部署在 Cruzr 机器人上,依赖于 WebSocket API,并使用 Python 实现。它支持各种机器人功能,例如文本到语音、语音到文本、导航、内容显示和扫描条形码。该系统的架构允许将接口移植到其他机器人和平台,展示其适应性。已经证明,中间件可以在 Pepper 机器人上运行,尽管尚未实现所有功能。中间件用于实现医疗保健用例,并收到了良好的反馈。

结论

讨论了中间件的云服务和本地语音服务需求,以便在不更改其他机器人上任何代码的情况下运行。给出了如何通过使用自然文本到代码生成器进一步简化编程接口的展望。对于使用上述平台(Cruzr、Pepper)的其他研究人员,可以使用新的中间件来测试人机交互。它可以用于教学环境,也可以适应使用相同接口和简单方法理念的其他机器人。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/3bba/10337241/9cdba5f243f3/12911_2023_2210_Fig1_HTML.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验