Using AI to monitor the impact of Covid-19 on our roads

0
Peter Mildon, COO and co-founder of Vivacity Labs, has been reviewing data from Vivacity’s national network of AI-based video road sensors to assess the impact of Covid-19 on UK highway networks on a daily basis. Here he considers what the long-term impact of Covid-19 will be on the UK’s transport habits in light of the climate emergency.
When we founded Vivacity in late 2015, one of our objectives was to make a positive impact to the way people used the roads in the UK. Our initial focus was on developing a cyclist sensor capable of operating on a truly multi-modal road space, in the hope that it would be used to encourage the modal shift away from polluting vehicles towards active travel.
Almost as a by-product of needing to positively identify other modes in order to differentiate them from cyclists, our sensor diversified into the full multi-modal sensor it is today. In 2016, we won Highways UK’s Intelligent Infrastructure Hub competition for the new technology most likely to revolutionise the transport industry.
At the time, I never imagined that the sensor network we had started to grow would prove so useful in helping at a time of national crisis, and less still that it would be used to monitor the impact government messaging aimed at reducing pedestrian and cyclist numbers in urban areas.
Over the past two weeks, we have been monitoring the impact on road usage during the Covid-19 outbreak. By Wednesday 25 March, there had been a 60% reduction in traffic across the UK. While some regions saw a quicker reduction in traffic numbers that others last week, since the lockdown on Monday 23 March, this reduction has been remarkably uniform from city to city, and from urban area to highway. Analysing the change by mode also yielded some interesting results:
  • Pedestrians saw the largest drop off at 80%
  • Car traffic dropped by 60%
  • Cyclists initially dropped very slowly, but since Monday have now dropped by 75%
  • Light Goods vehicles dropped by 45%
  • Initially there was no change seen in Motorbike or HGV volumes, but since Monday they have dropped by 65% and 40% respectively
  • Even buses have now dropped by 40% since the lockdown, indicating significant reductions in public transport services
The results clearly presented a drop in total traffic, but also a modal shift towards home deliveries.
Michael Vardi from Valerann, the 2017 Intelligent Infrastructure Hub winner, has reported similar traffic level drops from Israel and Spain, where Valerann’s Smart Road System is also monitoring traffic movements.
In Oxfordshire, we decided to push the analysis further. Our sensor network here has not just been recording the volume of traffic, but also the path that each road user was taking across the space. We decided to post-process the data to calculate if social distancing measures were being followed.

Example of measurement between pedestrians calculated with pedestrians coming within 2m of each other counted from stock footage. Red shows <2m; Yellow 2-3m; and green >3m

It has been discussed a lot in the media recently whether the government should be using mobile data to monitor social distancing. Many people are concerned that such an invasion of civil liberties might not be un-done once the crisis was over. This type of video analysis provides an alternative, which is non-invasive from a privacy perspective and offers a much higher resolution on the social distancing measurement. By using edge processing, no personal data is ever generated by the system – no videos are transmitted or stored, and the AI never knows who it saw. Instead we are able simply to acquire the data that is needed to help the Government make decisions on how its lockdown policy should adapt next.

Number of pedestrians passing within 2m of another pedestrian within the field of view of our 78 sensors in Oxfordshire per 5 minutes. Blue: Week commencing 8 March, Orange: Week commencing 15 March, Green: Week commencing 22 March. Data provided up to 24 March

The social distancing analysis showed that by 24 March, less than 48 hours after the Prime Minister announced a lockdown, peak daily pedestrian interactions had dropped by 70%, and the morning rush our peak was no longer disenable.

Clearly, Covid-19 is having a profound impact on all of our daily lives and has changed the approach to work for the majority of the UK work force. The question remains; what happens when all of this is over? Will everyone go back to their daily commute, or will companies finally realise that work can be done just as efficiently, and just as securely from home?
Given the climate emergency, I hope we don’t see road traffic demand bounce back to the ‘pre-Covid’ peaks. I was also reassured to see that demand for cycling initially held up compared with other non-delivery modes – perhaps this has encouraged some individuals to try cycling rather than take public transport, at least in the early stages of social distancing. When this is all over, hopefully some positives can grow from this major international crisis.
Click for more information on Vivacity Labs
Peter Mildon Co-founder of Vivacity Labs, is a member of Highways UK’s 2020 Intelligent Infrastructure Challenge steering committee. The Costain-sponsored Intelligent Infrastructure Challenge aims to accelerate the uptake of new ideas in the highways sector. This year’s competition will launch in late Spring. 
Highways UK takes place at the NEC Birmingham on 4/5 November. Click here for more information
Share.

About Author

Comments are closed.