Improving VTS radar tracking through software development



Kongsberg Norcontrol IT, Horten, Norway


The International Association of Lighthouse Authorities (IALA) V-128 Recommendations for Advanced Radars are aimed at improving the clarity of the radar image and track data for Vessel Traffic Service (VTS) operators, to ensure that an accurate, highresolution representation of the maritime domain and all targets within it, irrespective of size and speed, is available at all times.

However, much of today’s current radar hardware and software is simply not capable of presenting such a precise and accurate picture due to the noise and clutter returned by the radar signal. This is especially pertinent in today’s climate where small, fast moving targets are of much greater importance due to security concerns over possible terrorist threats.

Typical vessel tracking requirements for a port VTS radar may be to detect:

• Vessels in congestion, whose tracks may merge in close proximity

• Vessels that don’t follow assigned channels

• Vessels that enter restricted access areas. A coastal surveillance system must also be capable of detecting:

• Small, high speed targets that could be smugglers or people traffickers

• Boats with unusual moving patterns that could indicate illegal fishing or barter traders

• Boats that move slowly through radar clutter as they may be trying to avoid being seen.

In order to improve detection of such difficult targets and to enable high-resolution real-time vessel tracking for the VTS, Kongsberg Norcontrol IT employs Radar Extractor and Tracker technology within its VTS systems.

This is essentially the use of advanced software techniques to extrapolate an even clearer image from the raw radar signal. Radar Extractor and Tracker technology is already in use today but Kongsberg Norcontrol IT has recently developed the system even further in order to enable performance improvements by magnitudes.

The result is a new system integrated into Kongsberg Norcontrol IT’s C-Scope User Interface that offers nearly four times the sampling resolution available by current generation systems. These improvements enable the difficult targets to be displayed on an operators screen through little investment in hardware – it uses a high end PC and one COTS acquisition card, in addition to a generic radarinterface card.

Extracting the right track

It works thus: As the radar rotates and the extractor generates detections, the tracker receives the extracted signals and compares them against an internal track list. Any echoes that are consistent with the behaviour expected from the existing tracks are used to update the tracks, resulting in new positions and velocities. The remaining unassociated echoes are analysed and, if consistent over a number of scans, are used to generate new tracks. This function of association and dynamic filtering is fairly complex because radar echoes can be very variable and can also merge between nearby targets.

To determine what returns are consistent with an existing track, and to perform the track itself, the tracker uses a dynamic model that describes the potential behaviour of its tracks. The Radar Extractor and Tracker’s role is to extract and track the radar returns most likely to originate from vessels and other objects of interest. The main function of the extractor is to filter out all signals that represent noise, interference and clutter originating from the sea, rain, fog and reflections, leaving only signals that may be of potential interest. The tracker isolates from these detected signals those that are consistent with its tracking model and outputs a list of tracks. Each track represents a moving or stationary object and comes with a position and velocity in addition to other dynamical data.

The tracker must be able to initiate and correctly update tracks even when they are intermittent due to a low probability of detection, not lose them when they temporarily disappear due to shadowing, not confuse them when they are close to or overlapping with other echoes, and finally properly handle multiplicity when there is coverage overlap between nearby radars.

Cookie Policy. This website uses cookies to ensure you get the best experience on our website.