Machine Learning Giving Smarter Driving Advice

 Interesting how this is guided assistance, based on multiple goals.   Makes me recall using process maps to also drive goals.

Home/News/Using ML to Build Maps That Give Smarter Driving Advice/Full Text

ACM TECHNEWS

Using ML to Build Maps That Give Smarter Driving Advice

By MIT Technology Review. June 29, 2021

Scientists at Qatar's Hamad Bin Khalifa University (HBKU) applied machine learning (ML) to develop QARTA, a new automatic mapping service that can enhance traffic management with greater intelligence. HBKU's Rade Stanojevic and colleagues collaborated with the taxi firm Karwa to collect full global positioning system data on its fleet's comings and goings, so QARTA can advise routes for drivers at Karwa and other operators.

Stanojevic said QARTA's deeper understanding of actual road and traffic conditions in the city of Doha helps drivers shorten trips, translating into 5% to 10% greater fleet-wide efficiency.Stanojevic speculates that ML-based routing advice could factor into holistic views of cities, and help fleets slash carbon emissions by avoiding traffic jams.

From MIT Technology Review

Consider the Humble Roundabout

In my very early days of analytics, much involved with civil engineering data.    So this transcript of a Freakonomics article on the traffic roundabout was good.  These traffic management structures are often  seen in England, more rarely in the US.   But here locally getting more common.  This article contains traffic and accident statistics and arguments for and against, which I could have used back then.   Too late, but still interesting.

Quantum to Model Traffic with Simulations

Another mention of D-Wave quantum in the news.   Also the term 'Quantum-Style' is used for the first time I have seen,

Microsoft, Ford Try Using Quantum-Style Computing to Solve Seattle's Traffic Problem   By GeekWire via ACM

Microsoft and Ford Motor are using quantum-inspired computing models to try to optimize traffic management in Seattle.

Ford chief technology officer Ken Washington said timely optimization using an enormous number of possible route assignments is not feasible with traditional computers, so the partners have experimented with algorithms that simulate a quantum approach on classical systems.

The researchers tested various scenarios, including one involving about 5,000 vehicles concurrently requesting routes that spanned the Seattle area. Washington said the algorithms sent balanced routing suggestions to the vehicles in 20 seconds, improving congestion by 73% versus "selfish" routing, while cutting average commuting time by 8%.

Said Microsoft's Julie Love, “We don't have to wait until quantum computers are deployed on a wide scale to take advantage of the technology."  ... "

From GeekWire

 .... The results could be used to improve navigation apps, and could also be applied to optimization problems in fields ranging from robotics to aerodynamics.

Microsoft isn’t Ford’s only partner when it comes to traffic optimization: The automaker has also been working with NASA’s Quantum Artificial Intelligence Laboratory at Ames Research Center, where Burnaby, B.C.-based D-Wave Systems’ quantum annealing hardware comes into play. The NASA-Ford partnership focuses on using quantum-inspired algorithms to optimize energy consumption by commercial vehicle fleets.

Separately, Ford has set up a City Insights Platform to study urban mobility issues in depth. It’s all part of Ford’s drive to think of itself as a mobility company rather than strictly a car company, particularly as it closes in on fielding fully autonomous vehicles in 2021.  ... " 

Reducing Traffic with Reinforcement Learning

Our group looked at related problems in traffic and industry.    In Engineering.com

Machine Learning Techniques Aim to Reduce Traffic  by Michael Alba

It's a problem we can all relate to: sitting in traffic and waiting for a green light. While waiting, you may have even pondered how you would try to improve traffic efficiency—surely there's got to be some way for everyone to get to work on time.

But ponder no longer, because a team of engineers from Tsinghua University in China has handed the problem over to machines. The team’s recent study makes use of deep reinforcement learning algorithms to optimize traffic signaling, and its promising results suggest there may be a way to arrive on time after all.

Deep Reinforcement Learning

Let's be clear: traffic is a complex problem to solve, and traffic control engineers have long worked on improving efficiency. The difficulty arises because there are two distinct and challenging tasks involved—the first step is to create a useful model of traffic flow, and the next is to somehow find a way to optimize it.

The team modeled traffic flow using a simplified simulation of an eight-lane intersection, with only red and green lights (no yellows) and vehicles only allowed to go straight through (no right, left, or U-turns were permitted).

Using this simplified scenario, the team implemented reinforcement learning algorithms in order to determine signaling actions that were most beneficial to the system. This was evaluated by measuring the queuing length of traffic in both directions. By simulating different signaling situations, the algorithm aimed to minimize the length of traffic queues and therefore decrease driver wait time. ... " 

Inferring Traffic Patterns for Graph Analytics

I can see this kind of pattern recognition more broadly used, and linked not only to maps but also to network patterns for graph analytics.

Inferring urban travel patterns from cellphone data
Big-data analysis could give city planners timelier, more accurate alternatives to commuter surveys.
by Larry Hardesty | MIT News Office

In making decisions about infrastructure development and resource allocation, city planners rely on models of how people move through their cities, on foot, in cars, and on public transportation. Those models are largely based on surveys of residents’ travel habits.

But conducting surveys and analyzing their results is costly and time consuming: A city might go more than a decade between surveys. And even a broad survey will cover only a tiny fraction of a city’s population.

In the latest issue of the Proceedings of the National Academy of Sciences, researchers from MIT and Ford Motor Company describe a new computational system that uses cellphone location data to infer urban mobility patterns. Applying the system to six weeks of data from residents of the Boston area, the researchers were able to quickly assemble the kind of model of urban mobility patterns that typically takes years to build.

The system holds the promise of not only more accurate and timely data about urban mobility but the ability to quickly determine whether particular attempts to address cities’ transportation needs are working.  .... " 

Waze Making us Think Like Ants in Traffic

Ants avoiding traffic. Reminds me of swarming algorithms for tracking or searching.  Considered it in warehouses.   Have used Waze from time to time but have never had it direct me 'algorithmically'.    Doubt if many people do.  I do think this might be just the thing for self driving cars.    Once people trusted the swarm.

Smarter City Traffic Pilot

I notice in a current press release that the city of Cologne, Germany has completed a pilot of a smarter traffic system created by IBM.  In graduate school we used analytical prediction methods to address much smaller traffic problems, but even then the value was clear.  If the sensors and data can be economically positioned and the needed simulations done with that data gathered.  Traffic analysis is a great place to test the broader idea of many kinds of predictive analytics.  Good to see this being done to scale, will follow to see how it is ultimately put in production:

   " ... The City of Cologne, Germany, and IBM (NYSE: IBM) today announced the completion of a smarter traffic pilot to predict and manage traffic flow and road congestion in the city. The pilot demonstrates how the city of Cologne can anticipate, better manage, and in many cases, avoid traffic jams and trouble spots across the city using analytics technology. The city's traffic engineers and IBM were able to predict traffic volume and flow with over 90 percent accuracy up to 30 minutes in advance. As a result, travelers would be able to better plan ahead and determine whether they should leave at a different time, plan an alternate route or use a different mode of transportation. ... "