The Smart City:

The smartest street in the world

Advances in collecting and analysing data using next generation computing and human machine interaction are key to safer and more efficient urban streets.

Avoiding conflict on our busiest streets

Charles Chung (MMath '02) | Luis F. Miranda-Moreno (PhD '06)

Using multiple cameras mounted on vehicles and street corners, Brisk Synergies, founded by Charles Chung and Luis F. Miranda-Moreno, observe every interaction and near-collision between cars, buses, bikes and pedestrians. Using an automated video analysis of traffic flow, they measure how conflicts happen and advise municipalities on safer street design and policies.

Most often we recommend optimizing road design for lower speeds and better visibility. Everybody knows that speeding is dangerous, but people still do. It comes down to understanding human behavior and making the best recommendations to make the roads safer.

- Charles Chung (PhD '00), CEO Brisk Synergies 

Helping cyclists and motorists share the road

Kushal Mehta (MASc ’15) | Babak Mehran | Bruce Hellinga (BASc '89, MASc '90)

A research team from the University of Waterloo cycled hundreds of kilometres in Kitchener-Waterloo collecting data on the behaviour of motorists. Using sensors and handlebar cameras, they found that on roadways with bike lanes, unsafe passing by cars dropped to less than one per cent. Using that data, their algorithm-based virtual tool predicts the number of unsafe passes on roads, helping urban planners decide where bike lanes work best.

I think that living in a community in which people can choose from a range of safe and practical mobility options, including walking, biking, driving or taking public transit, increases our quality of life. But to achieve that, we need to improve infrastructure to reduce stress levels and improve safety for both motorists and cyclists.

— Bruce Hellinga (MASc '90), professor, civil and environmental engineering

Using processing power for traffic flow

Kurtis McBride (BASc '04, MASc '08) | Tony Brijpaul (BASc '04)

Understanding traffic is fundamental to managing the evolution of our cities. But, when Waterloo alumni Kurtis McBride and Tony Brijpaul looked at how cities were measuring traffic, they saw methods that hadn’t changed in decades – people with clipboards counting cars. They founded Miovision, a Kitchener-based company using computer vision, artificial intelligence and advanced modelling to make our intersections more efficient and safer.

The intersection is the foundation on which smart cities are being built. It’s a public space, distributed throughout the city, that already has infrastructure in place — the traffic light. Adding sensors, some processing power and connectivity to that intersection gives cities the power to better manage traffic. More importantly, it gives them the foundation for additional smart city applications.

— Kurtis McBride (MASc '08), CEO, Miovision

Learn more about Miovision

Fixing potholes with AI

John Zelek (BASc ’85) | David Chacra

Governments may soon be able to use artificial intelligence to easily and cheaply detect problems with roads, bridges and buildings. The Waterloo research team led by John Zelek, from Systems Design Engineering, analyzed photographs taken by the vehicle-mounted cameras used to create Google Street View to detect cracks and other defects. Often, municipalities will send teams out to identify potholes by eye. Zelek’s automated artificial intelligence system could cut costs, improve accuracy and lead to more timely repairs.

If governments have that information, they can better plan when to repair a particular road and do it at a lower cost. Essentially, it could mean lower taxes for residents. This system is a more consistent analysis because you’re not introducing the biases of different human beings who look at the data differently.

— John Zelek, professor, Systems Design Engineering

Adaptive traffic lights help drivers in snowstorms

Liping Fu | Zhengyang Lu (MASc ’16) | Tae J. Kwon 
(MASc ’11, PhD ’15)

Signals in modern cities are timed using optimization models that do not factor weather. By analyzing hours of video from a busy intersection, Liping Fu and his team from Waterloo’s Innovative Transportation System Solutions (iTSS) Lab simulated how traffic and driver behavior might change due to poor weather. Their recommendations, such as increasing the yellow interval at traffic lights, could prevent crashes, reduce delays by up to 20 per cent and better moderate traffic.

We need to have weather-responsive signal plans. Their timing should recognize weather conditions and change accordingly. The problem is that those parameters all assume normal weather conditions. In the winter, if the road surface is covered with snow and ice and visibility is poor, the numbers are not the same.

— Liping Fu, director, Innovative Transportation System Solutions (iTSS) Lab

geese working on computer screens