Skip Ribbon Commands
Skip to main content

Project title

Intelligent Computer Vision agents Optimising PTI Safety and Train Dwell Times (COF-PTI-02)

  • Project number:

  • Topic:

  • Status:


Project Summary

The project assesses the feasibility of improving PTI interaction by augmenting current CCTV technology with real time analysis of passenger movement and by correctly streaming passengers to the most appropriate part of the platform for their needs.

Project Abstract

Project Briefs are hosted on
The systems available to the drivers and platform staff can be improved and augmented such that PTI FWI incidents can be reduced, and PTI interactions can be optimised. In this proposal we suggest novel improvements to the current systems which tackle both the safety issue and go some way to optimising the PTI. To improve safety, the CCTV available to personnel will be augmented with an autonomous computer vision agent that detects and identifies potential PTI incidents. This benefits both the dispatch process including the DOO system and passenger safety; potential incidents are immediately highlighted, the driver has to actively discard the highlight before moving the train. Further, the system will interface with CCTV monitoring available to on platform staff such that they can also flag the potential incident and direct platform staff to investigate and assist.

In order to optimise the PTI interaction, we aim to detect the number of passengers and the available space on each carriage. In addition, we will identify the carriages that have facilities for PRMs and bicycles. To achieve this, modern advances in computer vision will be used applied to the CCTV cameras on board the trains. The information gathered by the cameras will be processed in real time and submitted to an intelligent agent (either centrally located or at the next platform). The system will use a series of on platform indicators (possibly floor LEDs) to show passengers where the quiet and busy carriages are, the door locations, and any gap / step issues in advance of the train arriving on the platform. The intention is to help guide passengers to the optimum boarding point

Read Project Brief

Project Reports

Project Reports are hosted on SPARK

This project has no reports available

This document is hosted on SPARK


You are now being redirected to SPARK, RSSB's knowledge sharing hub, to access the requested information.

Your RSSB user name and password can be used to access SPARK.