News

Last wave of truck counting cameras deployed in NSW

machine learning

The final batch of ‘machine learning’ truck counting and classification cameras in NSW are being rolled out in the state’s north and west.

Transport for NSW (Transport) currently has 64 machine learning traffic counting and classifying cameras in 23 key freight points across the state.

High-definition pictures of heavy vehicles are taken by the cameras, which are then classified into the type of vehicle in transit and the type of cargo being transported.

None is being used for enforcement, the agency stressed.

Transport said the aim of the trial is to “better understand freight movements to help reduce congestion, improve road safety outcomes, and encourage more efficient deliveries”.

The final 12 cameras will be installed at the following locations:

  • Castlereagh Hwy, Lightning Ridge
  • Carnarvon Hwy, Moree
  • Gwydir Hwy, Glen Innes
  • New England Hwy, Tenterfield
  • Bengalla Rd, Bengalla
  • John Renshaw Dr, Buttai
  • John Renshaw Dr, Buchanan

The cameras were first installed at Sheahan Bridge in 2022, along with bridge monitoring devices and the requirement for in-vehicle telematics and access conditions.

The bridge was limited to heavy vehicles up to 68.5 tonnes, which meant Higher Productivity Vehicles needed to detour to reach their destination.

“By using this technology, ongoing access for heavy vehicles has been increased to 85 tonnes,” said a Transport spokesperson.

“This is an example of how Transport is using this technology and data to improve access.”

The value of a unit may vary depending on the configuration, the spokesperson added. Installation and maintenance costs may also vary depending on location and the configuration of units installed.

The technology is seen as another tool that can be used to help manage increasing freight volumes, which are expected to increase across NSW by 28 per cent by 2036 over a 2018 baseline.

Leave a Reply

Your email address will not be published. Required fields are marked *

Send this to a friend