Walter, DJ; Groves, PD; Mason, RJ; Harrison, J; Woodward, J; Wright, P; (2013) Novel Environmental Features for Robust Multisensor Navigation. In: Proceedings of the 26th International Technical Meeting of The Satellite Division of the Institute of Navigation (ION GNSS 2013). (pp. 488 - 504). The Institute of Navigation: Manassas, US.

Preview

PDF ION_GNSS13_debbie_walter.pdf Available under License : See the attached licence file. Download (2MB)

Abstract

Many navigation techniques have now become so reliant on GNSS that there is no back up when there is limited or no signal reception. If there is interference, intentional or otherwise, with the signal, navigation could be lost or become misleading [1]. Other navigation techniques harness different technologies such as Wi-Fi [2], eLoran and inertial navigation. However, each of these techniques has its own limitations, such as coverage, degradation in urban areas or solution drift [3]. Therefore there is a need for new navigation and positioning techniques that may be integrated with GNSS to increase the reliability of the system as a whole. This paper presents the results of a feasibility study to identify a set of novel environmental features that could be used for navigation in the temporary absence of GNSS or degradation of the signal. By measuring these features during times of GNSS availability a map can be produced. This can be referred to during times of limited reception, a principle already used for some Wi-Fi positioning techniques [2]. Therefore a “measurable” can be defined as a feature either man-made or natural that is spatially distinct and has limited temporal variation. Possibilities considered include magnetic anomalies [4], light intensity and road signs. Firstly, a brainstorming exercise and a literature study were conducted to generate a list of possible environmental features that was assessed for the viability of each candidate. The features were ranked according to three criteria: practicality, precision and coverage. The definition of practicality for each measurable was that a suitable detector must be installable on a road vehicle, particularly an emergency vehicle, at a reasonable cost with minimal alterations to the vehicle. Precision was defined in terms of the spatial variation of the environmental feature and thus the accuracy with which position information might be derived from it. Coverage was assessed in terms of the availability of the feature over a range of different environments. Continuous coverage is not required because the new measurables may be used in combination and integrated with dead reckoning techniques, such as odometry and inertial navigation [3]. The outcome of the viability study was used to determine which features are to be experimentally tested. Magnetic anomalies, road texture and a dozen other environmental features were found to be worth investigation. Features which were discounted include wind speed and pulsars [5]. The initial experiment was carried out on foot in Central London. The same tests were repeated on two separate days, with a closed loop circuit walked three times on each occasion. This experiment used an Inertial Measurement Unit (IMU), comprising accelerometer and gyro triads, together with a barometer, three-axis magnetometer and GNSS receiver. The experiment was also recorded using a camcorder from the point of view of a pedestrian, enabling visual and audio features of the environment to be assessed. Magnetic anomalies were found to be a promising source of position information. Peaks in the magnetometer data were observed on all rounds at approximately the same positions. There were also similarities seen in the temperature profiles after correcting for the temporal variation of the background temperature. Another potential source of position information was found to be text-based signs. It is relatively simple to extract text from camera images and it is easily stored in a feature database. However, methods of dealing with identically-worded signs in close proximity will need to be developed. Sound levels were analysed in 10s intervals for the mean, minimum and maximum sound volume. There was no clear correlation observed between the different rounds of the experiment. Due to the pedestrian experimental results sound levels of the surroundings will not be used in further experimentation. An alternative area of enquiry for using sound (in the vehicular experiments) is using microphones to indirectly measure road texture based on the noise from the wheel contact with the road [6]. The paper will also present results of road vehicle experiments. Multiple circuits of the same routes will be compared. Different environments will be assessed including rural, dual carriageways, suburban and urban roads. Sensors to be used include the IMU and 3-axis magnetometer from the pedestrian experiment, a barometer, gas sensors, a microphone, an axle-mounted accelerometer, an ambient light sensor and a thermometer. These will be placed either on, inside or under the vehicle as determined by the individual needs of the sensors. The results will be used to determine which of these sensors could be potentially used for a multisensor integrated navigation system and also the environments in which they work optimally. Using the results of the three feasibility study phases (literature review, pedestrian and road experiment) the next project stage will be to produce a demonstration system that uses the most feasible features of the environment and creates a map database during times GNSS is present. This database will then be used for navigation in times of need. In the long term, it is envisaged that this technique will be implemented cooperatively, with a batch of vehicles collecting feature data and contributing it to a common shared database. / References [1] Thomas, M., et al., Global Navigation Space Systems: Reliance and Vulnerabilities, London, UK: Royal Academy of Engineering, 2011. [2] Jones, K., L. Liu, and F. Alizadeh-Shabdiz, “Improving Wireless Positioning with Look-ahead Map-Matching,” Proc. MobiQuitous 2007, Phildaelphia, PA, February 2008, pp. 1-8. [3] Groves, P.D., Principles of GNSS, Inertial, and Multisensor Intergrated Navigation Systems, Second Edition, Artech House, 2013. [4] Judd, T., and T. Vu, “Use of a New Pedometric Dead Reckoning Module in GPS Denied Environments,” Proc. IEEE/ION PLANS, Monterey, CA, May 2008, pp. 120?128. [5] Walter, D. J., "Feasibility study of novel environmental feature mapping to bridge GNSS outage," Young Navigator Conference, London, 2012. [6] Mircea, M., et al., “Strategic mapping of the ambient noise produced by road traffic, accordingly to European regulations,” Proc. IEEE International Conference on Automation, Quality and Testing, Robotics, Cluj Napoca, Romania, May 2008.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item