2022
08/05
相关创新主体

创新背景

GPS导航系统为人们的出行以及一些探测工作带来了极大便利,但对环境变化缺乏一定的感应,需要不断精进,提高导航的精准性,扩大使用范围。

 

创新过程

韩国科学技术学院(KAIST)计算学院的韩东洙教授带领智能服务集成实验室的研究人员开发了一种集成AI 的GPS系统,可在室内或室外任何环境下进行高质量的精准工作,开拓了定位技术的运用。
基于AI的室内/室外集成GPS系统或IOI GPS系统除了使用室外GPS信号,还会使用来自压力传感器、惯性传感器、地磁传感器和光传感器等来自多处的信号,提高对室内位置的估测精准度。GPS定位导航需要网络信号的支持,在无线局域网信号或基站信号无法到达的空间无法估计位置并进行导航。新的IOI GPS突破这一限制,在没有信号或室内建筑物中也能实现定位。

研究团队为新系统开发了检测环境变化的技术,当进入建筑或底层、楼梯、电梯等,人工智能可以检测建筑层数,随着环境变化改变定位方式。新系统结合各种地标检测技术和行人惯性导航工具PDR,设计出传感器融合定位算法。
新算法在建筑物内可以提供准确的楼层信息。无需像地磁定位技术或无线LAN的定位方法一样依赖视觉数据,不需要任何事先准备就能在世界上任何地方的室内和室外工作使用。
研究人员为IOI GPS系统生产了一种便于操作的电路板。电路板中装有接受和处理GPS的芯片、WIFI、蓝牙信号以及惯性传感器、气压计、磁力计和光传感器,包括研究新开发的传感器融合定位算法。

进行IOI GPS板的精度实验测试时,地板和距离的估计精度分别为95%和3至6米的误差;导航模式改变室内/室外模式需要0.3秒的过渡时间。结合PDR技术使用时,系统的估计精度有所提高,误差范围减小到一米。
研究致力于组装一个带有内置定位板的标签,可用于追踪儿童和老年人,定位在灾难或危险场所迷路的人或救援人员,以及应用在博物馆、科学中心和美术馆,就访客所在的当下位置进行讲解。并且,研究人员一直在探索可用于车辆的传感器融合定位算法和定位板,方便追寻进入地下停车场等室内区域的车辆。

研究人员表示,室内/室外集成GPS系统可以精确定位建筑物中没有无线信号或室内地图的位置,并且可以应用于无限多个区域。当用于车辆的IOI GPS板制造出来时,研究团队将与汽车制造商和汽车租赁公司合作,还将为智能手机开发传感器融合定位算法。

 

创新关键点

基于人工智能的强大功能开发室内/室外集成GPS系统,突破网络信号以及环境变化对GPS定位导航的限制,扩展GPS的应用领域。

 

Innovate GPS positioning technology with AI

Professor Han Dong-so of the School of Computing at the Korea Institute of Science and Technology (KAIST) led researchers in the Intelligent Service Integration Laboratory to develop a GPS system with integrated AI that can perform high-quality precision work in any environment, indoors or outdoors, pioneering the use of positioning technology.

Ai-based indoor/outdoor integrated GPS systems or IOI GPS systems use signals from multiple locations, such as pressure sensors, inertial sensors, geomagnetic sensors, and light sensors, to improve the accuracy of estimating indoor positions. GPS positioning and navigation requires the support of network signals, and it is impossible to estimate location and navigate in spaces where wireless LAN signals or base station signals cannot reach. The new IOI GPS breaks through this limitation and enables positioning even in buildings without a signal or indoors.

The research team has developed technology for the new system to detect environmental changes, when entering the building or the ground floor, stairs, elevators, etc., artificial intelligence can detect the number of floors of the building, and change the positioning method as the environment changes. The new system combines various landmark detection technologies with pedestrian inertial navigation tool PDR to design a sensor fusion positioning algorithm.

The new algorithm can provide accurate floor information within a building. There is no need to rely on visual data like geomagnetic positioning technology or wireless LAN positioning methods, and it can be used indoors and outdoors anywhere in the world without any prior preparation.

The researchers produced a circuit board for the IOI GPS system that was easy to operate. The circuit board is equipped with chips that receive and process GPS, WIFI, Bluetooth signals, as well as inertial sensors, barometers, magnetometers and light sensors, including research on newly developed sensor fusion positioning algorithms.

When the accuracy experimental test of the IOI GPS board is carried out, the estimated accuracy of the floor and distance is 95% and the error of 3 to 6 meters, respectively; Navigating mode To change the indoor/outdoor mode takes 0.3 seconds of transition time. When used in conjunction with PDR technology, the estimation accuracy of the system is improved, and the margin of error is reduced to one meter.

The research is focused on assembling a tag with a built-in positioning plate that can be used to track children and the elderly, locate people or rescuers who are lost in disaster or danger places, and apply it to museums, science centers, and art galleries to explain where visitors are. Moreover, researchers have been exploring sensor fusion positioning algorithms and positioning plates that can be used in vehicles to facilitate the pursuit of vehicles entering indoor areas such as underground parking lots.

The researchers say that the indoor/outdoor integrated GPS system can pinpoint locations in buildings where there is no wireless signal or indoor map, and can be applied to an infinite number of areas. When the IOI GPS board for vehicles is manufactured, the research team will work with automakers and car rental companies, and will also develop sensor fusion positioning algorithms for smartphones.

智能推荐

  • 物联网+测控技术与仪器 | 新型低功耗辐射剂量计可实时监测现场

    2022-11-22

    AIST的研究人员开发的这种使用节能无线技术的紧凑型辐射剂量计。两年或更长时间内不需要更换电池。

    涉及学科
    涉及领域
    研究方向
  • 通往天空的大海—韩国公寓塔楼

    2022-07-12

    这是一幢由Maaps Architects设计的,建于2017年的韩国公寓。该项目的设计灵感来源于海洋,被称为“通往天空的大海”。玻璃的建筑立面倒映着天空的蔚蓝,或封闭或开放的阳台形成错落的视觉,象征着波光粼粼的海面,建筑底层的白色几何结构的外廊柱就像是海浪拍打在沙滩形成的泡沫。这栋公寓建筑独特的设计,使得大众对都市建筑有了新的认识,体现了人类对自然的向往。

    涉及学科
    涉及领域
    研究方向
  • 机器人+海洋科学 | 利用“追踪海上机器人”研究幼虫

    2022-08-24

    加州大学戴维斯分校的科学家们正在旧金山北部的博德加湾(Bodega Bay)的海洋中部署“机器人幼虫”,类似“小黄人”的机器人揭示了海洋幼虫惊人的移动方式。

    涉及学科
    涉及领域
    研究方向
  • 新材料 | 利用“焦耳热闪蒸”节能技术将汽车废塑料变成石墨烯

    2022-07-01

    美国莱斯大学研究团队利用焦耳热闪蒸技术,使用电流加热碳,将二手车塑料转换成高质量石墨烯,为车辆产生的垃圾提供了一个潜在处理办法,并为环保生产石墨烯提供了新思路。

    涉及学科
    涉及领域
    研究方向