Intelligent Geomatics Systems for Reliable Positioning in Smart Cities

点击翻译按钮获取中文摘要

Global Navigation Satellite Systems (GNSS) provide positioning to a broad range of users operating in outdoors however indoor and urban navigation remains challenging in terms of availability, accuracy and reliability. The majority of GNSS users are now individuals with wireless handsets and that spend the majority of their time in urban and indoor environments. New ground-based wireless communication standards that support direct range measurement, including Ultra-Wideband (UWB), 802.11mc (WiFi) and 5G are emerging while Bluetooth Low Energy (BLE) is becoming the standard for low precision ranging and contact tracing. The objective of this research is to improve indoor and urban navigation to the several centimetres level through integration of GNSS with these ranging systems. The proposed research aims to: 1. Integrate new ranging technologies with GNSS in urban environments. The performance of IEEE 802.11mc and 5G ranging will be investigated and compared with UWB ranging and improved error models for 802.11mc and 5G will be developed. Artificial intelligence will be applied to correct measurement errors and the results integrated with GNSS observations to estimate position and each system's systematic errors. 2. Improve BLE ranging using 802.11mc, 5G and UWB. Low power, low cost, low complexity BLE has been identified as a favoured method for electronic contact tracing in the context of the 2020 Covid-19 pandemic. While BLE can determine if devices have been in proximity, its use for range estimation is limited. Recent work using artificial intelligence techniques to learn to identify changes in BLE observations due to obstructions will be expanded to develop a concept where a subset of devices equipped with the other 3 ranging systems can be used to train low-cost systems that rely only on BLE. 3. Integrate GNSS and ranging for maximum availability, accuracy and reliability in urban and indoor environments. The statistical reliability of integrating limited GNSS with each wireless ranging systems in 1 and 2 will be evaluated. Models for reliability will be developed for autonomous driving, pedestrian navigation and contact tracing. New models for both the systematic and stochastic errors of each of the systems will be developed and the advantages and disadvantages of including each type of measurement in different user scenarios will be assessed. This work is significant because GNSS is not sufficiently available in urban and indoor environments, nor reliable for safety-of-life applications. Current Bluetooth-based ranging is also not accurate enough for reliable contact tracing. The reliability of both technologies will be improved if appropriately integrated with the three emerging wireless ranging technologies (802.11mc, 5G and UWB) targeted in this proposal. The results will be directly useful to Canadian and international automobile manufacturers, health authorities, telecommunications companies, and location-based services providers.