Apps, smart traffic signals complement EVs to save energy, commute time

Waze, the community-based traffic and navigation app, aims to speed you on your way to your destination with real-time input on traffic conditions from other drivers. Waze seems like such a good idea that Google bought the Israel-based company for an estimated $1 billion (prompting investigations by the UK Office of Fair Trading and the US Federal Trade Commission).

Of course, Waze would seem to speed you and fellow Waze users around traffic jams and the expense of the non-Waze users stuck in those jams. It suggests an arms race of sorts—if everyone is Waze-equipped, it seems no one would benefit.

Here is what Waze has to say: “Imagine 30 million drivers out on the roads, working together towards a common goal: to outsmart traffic and get everyone the best route to work and back, every day.” But really, you're not outsmarting traffic—you are outsmarting the people who don't use Waze.

Of course, if Waze offers you significant advantages, by all means adopt it. However, other approaches might provide more general benefits. One technology is described in a paper published in the September 2013 IEEE Transactions on Intelligent Transportation Systems by S. El-Tantawy of the Intelligent Transportation Systems Center and Testbed, University of Toronto, and coauthors B. Abdulhai and H. Abdelgawad. The paper describes an adaptive traffic-signal control (ATSC) system that “has shown strong potential to effectively alleviate urban traffic congestion by adjusting signal timing plans in real-time in response to traffic fluctuations…. Efficient and robust ATSC can be designed using a multiagent reinforcement learning (MARL) approach in which each controller (agent) is responsible for the control of traffic lights around a single traffic junction.”

The authors have integrated ATSC and MARL to develop multiagent reinforcement learning for integrated network of adaptive traffic signal controllers (MARLIN-ATSC), which has been tested on a simulated network of 59 intersections in Toronto during morning rush hour, reducing average intersection delays from 15% to 39%.

Writing in Slate, Jeffrey Ball describes such technologies as constituting what some call the “smart road”—the pavement analog to the smart grid (although “smart road” in this sense should not be confused with the Virginia Tech Transportation Institute's Virginia Smart Road).

Ball includes in the smart-road category projects like Safe Road Trains for the Environment (SARTRE), funded by the European Commission, which aims to allow vehicle platoons to operate on normal public highways. With SARTRE, a lead vehicle operated by a professional driver will be followed by closely spaced semiautonomous vehicles whose nominal “drivers” can read, eat, talk, or watch a movie.

I would put SARTRE in the smart-vehicle category, not smart road. However, Ball's larger point is correct when he suggests that apps like Waze and technologies like MARLIN-ATSC or SARTRE aren't as flashy as electric vehicles. Waze on your mobile device won't impress your neighbors the way a Tesla S in your driveway would.

Nevertheless, Ball suggests, Waze and adaptive traffic-signal controllers may promise to do more to cut energy use than low-volume electric cars. “Cool technology comes in many varieties,” he concludes. “For the foreseeable future, the kind that’s largely invisible may matter more than the kind that’s shiny and red.”