Toyota, Turing Institute Focus on Marrying Mobility & AI

The Toyota Mobility Foundation and the Alan Turing Institute are partnering to transform the way cities approach mobility with the help of artificial intelligence and autonomous vehicles.

Spanning 18 months, the collaboration will bring together mobility specialists from Toyota and software engineers and other researchers from the Turing and the Universities of Cambridge and Manchester.

The goal of the partnership is to transform complex traffic management from static systems into those managed in real-time across many types of mobility — public, private and hybrid forms.

The project emerged out of the realization that data science and AI can contribute to delivering optimized flows of traffic within cities, Nicolas Guernion, the Turing Institute’s director of partnerships, told The Connected Car.

“Researchers will be developing a visualization platform to empower city planners in simulating changing behavior patterns and thereby optimize their decisions for traffic control, particularly how to manage traffic signals,” Guernion explained.

The project will focus on the creation of graphic interfaces allowing city planners to analyze traffic flows and test out scenarios from manipulating traffic lights.

This will be underpinned by developing algorithms and integrating machine learning models to predict traffic flows.

To inform the modeling, the project will look at the impact of different scenarios for future traffic planning, such as a decrease in individual traffic participants, increase in shared riding or a rise in the share of fleet vehicles.

“Ultimately, the project will help smooth traffic flows which reduces commuting time and pollution within cities,” Guernion said. “It will also help strengthen the resilience of the traffic management system.”

Dr. William Chernicoff, senior manager of global research and innovation at Toyota Mobility Foundation, told The Connected Car that autonomous and connected vehicles are bound to play a substantial role in future cities.

“Personal mobility will continue to play an important, and possibly expanded role in enabling all people, regardless of ability or status, to move freely and have the opportunity to achieve their potential,” he said.

Chernicoff explained AVs will not just be connected to each other, but also the infrastructure around them.

Potential outcomes from the collaboration could include integrating an AI system for traffic lights and other signal controls, building a platform to monitor and predict traffic behavior, and to test out planning scenarios by sharing data about congestion or pollution hotspots.

“The ability of the infrastructure to effectively manage these vehicles and other mobility modes will dictate their collective value and extent of the role they can play,” Chernicoff said.

As with many things in the digital world, the biggest stumbling blocks for smart cities will continue to shift over time, especially as computing power continues to improve, Chernicoff added.

“For the time being, one of the challenges for the near and foreseeable future will be the ability to quickly process data so that it allows for timely positive impact,” he explained.

Chernicoff explained it is not just about computational power — it’s also about the underlying data science and mechanics of the machine learning.

“The available data must consistently, reliably, accurately, and with reasonably precision, be processed, analyzed, presented, and acted upon within seconds,” he said.