Traffic Ear

Optimizing traffic management through advanced acoustic analytics

Challenge

Congestion has increased rapidly in recent years at a heavy cost to the UK economy. It has led to a significant reduction in productivity levels and stalled economic growth. Moreover, congestion also poses a health hazard due to its significantly high contribution of carbon emissions. Unfortunately, most of the solutions in the market gather a very limited range of data points and are unable to provide effective, robust and dynamic traffic management solutions. Therefore, there is a great need for more advanced and optimized traffic management systems.

Solution

We have developed a low-cost acoustic sensor that detects vehicle movement using tyre noise within city and motorway environments by utilising our deep learning algorithms, highly detailed data is captured to:

  1. Detect the number and size of vehicles (e.g. car, bus, lorry, motorcycle) within an urban environment, analysing traffic movements and congestion causes.
  2. Detect vehicle emission levels by analysing both size of the vehicle and the engine noise to identify engine type (Petrol/Diesel/Electric/Hybrid). We can determine an estimation of vehicle emissions based on novel acoustic AI models that have been trained on both vehicle noise and air pollution data with Spatiotemporal Context.

This sensor will further tune the existing algorithms in a greater number of locations and test whether or not it is possible to predict traffic levels, vehicle emission and noise levels.

Where To Use

Cities

Clean zone

Construction sites

Motorway

Schools