by Martin Litzenberger, Carmina Coronel and Christoph Wiesmeyr (AIT Austrian Institute of Technology GmbH)
Real-time traffic situation monitoring is key to optimising traffic flow using intelligent traffic systems (ITS), especially in the urban environment. Often, the high density of traffic sensors needed to achieve a seamless real-time monitoring of important arterial roads is difficult to implement due to technical constraints or installation costs. Fibre optic acoustic sensing (FOAS) is a relatively new technology that allows a seamless, real-time monitoring of road traffic over distances of up to 50 km, using the existing telecom fibre-optic cable infrastructure.
Roads have always been the backbone of transportation in the urban environment, and continual traffic monitoring is crucial to ensure continuous traffic flow. The data provided by real-time road traffic monitoring can potentially provide information about traffic jams and accidents. This information can help traffic management centres to react quickly to incidents, and intelligent transportation system (ITS) measures, such as the closure of a lane or temporary usage of the hard shoulder, can automatically be implemented.
Different technologies are currently used for traffic monitoring systems where sensors are either installed overhead, under, or next to the road to monitor traffic flow [1]. Such sensors could be laser scanners, infrared, radar, ultrasonic, magnetic, acoustic or video cameras. Another method for traffic monitoring is through crowd-sourcing of smartphone connection data or from fleets of vehicles equipped with GPS systems ("floating car") [2]. Sensors installed under the road surface come with the disadvantage of the high costs of ongoing repairs and maintenance while sensors placed overhead or next to the road, such as cameras, are susceptible to adverse weather conditions.
FOAS is a technology that allows seamless, real-time monitoring of traffic over distances of up to 50 km without additional roadside installations [3]. It uses fibre-optic cables that are already installed in the ground next to the roads for data- and communication-networks (telephone, internet), as a distributed detector. The advantage is that the fibre-optic cable infrastructure typically installed at high density in the urban environment can be used, as it is, for traffic sensing by connecting an optical "interrogator" instrument to one end of an unused fibre. The technique allows the detection of very small changes in the optical fibre, such as the mechanical strain caused by microscopic deformations from vibrations of the cars running nearby.
FOAS systems work by sending short laser pulses through a fibre-optic cable where the light is scattered via Rayleigh scattering and the light returning to the source is analysed. In FOAS systems, optical fibres of up to 50 kilometres in length can be used. An interrogator device connected to one end of the fibre transmits a series of laser light pulses into the optical fibre, as shown in the top of Figure 1.
Figure 1: Top: Principle of the FOAS measurement for traffic situation monitoring. Bottom: Typical dataset of an FOAS spectral power diagram after thresholding showing recorded vehicle trajectories over 5 minutes and 4.8 km distance.
In the glass of the optical fibre, an effect occurs that causes a continuous "reflection" of the light along the fibre. Rayleigh scattering is caused by inhomogeneities in the glass and is actually a different mechanism than reflection, but for the sake of simplicity, the Rayleigh scattering effect can be described as light being reflected on a myriad of microscopic mirrors embedded in the glass. Therefore, for a single laser pulse being coupled into the fibre, instead of many distinct reflected pulses, a continuously distributed signal is returned from the fibre. The scattered light has the same frequency as the impinging light wave and can be analysed by optical means. The vibrations generated by the passing cars and trucks stretch and compress the optical fibre, affecting its optical path length. This induces a measurable phase shift in the back-scattered light, which is sensed by interferometric methods. Probing the fibre with a laser pulse of high repetition frequency (2 kHz) makes it possible to analyse the vibration spectrum produced by nearby vehicles, distinguishing them from other vibration sources and tracking their time-location trajectories along the cable.
We have recorded FOAS data for traffic flow over two days from a road section of 20.4 km in length, with the fibre-optic cable being installed next to a highway in the urban region of Graz, Austria. After thresholding of the spectral power diagram of the raw FOAS signal we obtain an “image” representation of the vehicle tracks as a time-location diagram, in which the vehicles running on the road can been identified. Figure 1 (bottom) depicts an example of the time-location diagram for five minutes duration and a road section of 4.8 km length, recorded at 2.4 km distance from the interrogator device. Driving direction and vehicle speed are represented by the inclination of the tracks in the diagram. The vehicles driving on the carriageway closer to the optical fibre cable are well represented in the data, while the vehicles driving on the other carriageway, in the opposite direction, cannot be resolved as well. A real-time estimation of the traffic situation by automatically detecting the speed changes from the trajectory pattern seems feasible, at least for the closer lanes. Traffic density can be derived as well, from the line density in the diagram. The project "Roadwise" funded in the EUROSTARS framework of the European Union, is currently investigating such methods for automatic, real-time and seamless detection of the traffic situation from FOAS data.
Given that FOAS systems only require the installation of an interrogator device connected to one end of an existing fibre-optic cable, the presented solution requires low-cost roadside maintenance and installation. An additional advantage of a FOAS-based traffic situation monitoring system is its long-range capabilities and seamless nature of the measurement.
References:
[1] J. Guerrero-Ibáñez, S. Zeadally, and J. Contreras-Castillo: “Sensor Technologies for Intelligent Transportation Systems”, Sensors (Basel), vol. 18, no. 4, Apr. 2018, doi: 10.3390/s18041212.
[2] V. Astarita, et al.: “Floating Car Data Adaptive Traffic Signals: A Description of the First Real-Time Experiment with ‘Connected’ Vehicles”, Electronics, vol. 9, no. 1, Art. no. 1, Jan. 2020, doi: 10.3390/electronics9010114.
[3] E. Catalano, et al.: “Automatic traffic monitoring by ϕ-OTDR data and Hough transform in a real-field environment”, Appl. Opt., AO, vol. 60, no. 13, pp. 3579–3584, May 2021, doi: 10.1364/AO.422385.
Please contact:
Martin Litzenberger
AIT Austrian Institute of Technology GmbH, Austria
