Page 28 - 卫星导航2021年第1-2合期

El‑Sheimy and Li Satell Navig (2021) 2:7 Page 18 of 23

Fig. 10 Photodiode/Camera integrated navigation (modifed on the results reported in Zhuang et al. (2019))

Multi-platform, multi-device, and multi-sensor information cooperative localization can be an important feature for
fusion 5G networks. In the coming years, the characteristics
Te PLAN system will develop towards the integration of of massive devices, dense base stations, and device-to-
multiple platforms, multiple devices, and multiple sen- device communication may make accurate cooperative
sors. Figure 11 shows a schematic diagram of the multi- localization possible. In addition to multiple devices,
ple-platform integrated PLAN. there may be multiple devices (e.g., smartphones, smart-
With the development of low-cost miniaturized satel- watches, and IoT devices) on the same human body or
lites and Low Earth Orbit (LEO) satellite technologies, vehicle. Te information from such devices can also be
using LEO satellites to provide space-based navigation used to enhance PLAN.
signal has become feasible. Te research paper (Cluzel
et al. 2018) uses LEO satellites to enhance the coverage Self-learning algorithms and systems
of IoT signals. Also, the paper (Wang et al. 2018) analyzes Artifcial intelligence
the navigation signals from LEO satellites. In addition to With the popularization of IoT and location-based ser-
the space-borne platform, there are airborne and under- vices, more complex and new PLAN scenarios will
ground PLAN platforms. For example, the research paper appear. If this is the case, self-learning PLAN algorithms
(Sallouha et al. 2018) uses unmanned aerial vehicles as and systems are needed. Tere are already research
base stations to enhance PLAN. works that use artifcial intelligence techniques in various
Collaborative PLAN is also a future direction. Te PLAN modules, such as initialization, the switch of sen-
research in (Zhang et al. 2017a) has reviewed 5G coop- sor integration mode, and the tuning of parameters. Te
erative localization techniques and pointed out that research paper (Chen et al. 2020) uses ANN to generate