A fairy tale approach to cooperative vehicle positioningTools Stephenson, Scott, Meng, Xiaolin, Moore, Terry, Baxendale, Anthony and Edwards, Tim (2014) A fairy tale approach to cooperative vehicle positioning. In: 2014 International Technical Meeting of The Institute of Navigation, 27-29 Jan 2014, San Diego, California. Full text not available from this repository.AbstractThis paper outlines an innovative approach to the cooper-ative positioning of road vehicles by sharing GNSS informa-tion. Much like the children’s fairy tale Hanzel and Gretel by the Brothers Grimm, GNSS receivers on road vehicles generate detailed VRS-like “breadcrumbs” as they accurately position themselves (in this case using a Network RTK GNSS technique). These breadcrumbs can then be shared with other vehicles in the locality to help position themselves, much like traditional RTK GNSS positioning. Similar to the breadcrumbs in the fairy tale that are eaten by birds shortly after being dropped, the VRS-like correction information is only valid for a short period of time. By using this technique, off-the-shelf GNSS receivers can be used without any major hardware or software adjustments, including those of different receiver brands or legacy receivers. The techniques employed in this paper aim to deliver absolute positions, to enable high-accuracy ITS applications that involve road agents and infrastructure alike. A much anticipated development in ITS technology is the use of vehicle to vehicle or vehicle to infrastructure commu-nication (collectively called V2X). Driven partly by the need to increase road safety, and perhaps heavily influenced by the infotainment needs of drivers and passengers, V2X technology will allow local vehicles to communicate with each other and with other road agents and fixed infrastructure. In the US, the National Highway Traffic and Safety Administration (NHTSA) recently commented that connected vehicle technology “can transform the nation’s surface transportation safety, mobility and environmental performance”, with industry experts pre¬dicting the widespread uptake of the technology within 5-6 years. This provides an opportunity for road vehicles to share GNSS information. (As the V2X technology is not under test in this paper, any V2X communication is made using a local Wi-Fi P2P network). This is demonstrated in this paper by directly sharing Network RTK correction information for one receiver (in this case Virtual Reference Station (VRS) corrections) with a second receiver on a separate vehicle. This is done using an NTRIP client running on an Android cellular device at the end-user distributing the VRS corrections from the NTRIP server to both the primary and secondary receivers (in the same locality). Network RTK corrections are not always available, not least because it requires a subscription to a service provider. However, if a GNSS receiver on a road vehicle has access to raw GNSS observations and is capable of calculating its absolute position to a reasonable accuracy (perhaps using an integrated sensor approach), then it has the necessary ingredients to generate its own VRS-like RTK corrections. These VRSs are left like breadcrumbs in the road, ready for any other GNSS receiver in the vicinity to use. Any received VRS correction information will continue to be valid for up to 10 seconds. By utilising the open source RTKLIB GNSS processing software, and the most recent RTCM standard messages (RTCM v3.1) generated through software provided by BKG, one receiver can perform the task of a VRS or a moving base station. The position of the receiver is processed whilst separately recording the raw RINEX information, in order to generate an RTCM stream that simulates that of a Network RTK VRS correction service. Additional information about the source of the correction information is also transmitted, in-cluding the self-assessed quality of the position and hardware used, using the RTCM message types reserved for proprietary information from service providers. Sharing GNSS information between vehicles is shown to significantly increase the availability of ambiguity fixed so-lutions, for both dual and single frequency receivers; and improves the performance of DGNSS receivers. However there needs to be caution, as the use of a single epoch of raw observations from a moving base station is less reliable than traditional static base station Network RTK GNSS positioning. Fixing the integer ambiguity is more likely to be successful (passing the ratio test), but also more likely to be incorrect, and relies heavily on the initial position of the moving base station (i.e. the relative position or baseline may be accurate, but not necessarily the absolute position). Three control solutions are used to assess the performance of the cooperative positioning techniques in real world tests: An RTK GNSS control solution provided by a local static continuously operating reference station (CORS); a Network RTK GNSS solution based on the MAC standard; and an Applanix POS/RS dual frequency GPS inertial navigation system. The processing parameters are adjusted to assess the optimum configuration for successful cooperative positioning (delivering accuracy and reliability), and the limitations of the technique are addressed. It is shown that although the cooperative position may not match the positioning accuracy of the initial moving base station vehicle (<5 centimetre), the solution is valid for sub-decimetre accuracy for up to one minute using dual frequency GPS observations. A cooperative DGNSS solution is accurate to 20 centimetres over the same period.
Actions (Archive Staff Only)
|
Tools
Tools
