image title
Seen Google camera backpacks and self-driving Uber cars on campus? Here's why.
March 10, 2017

In separate projects, tech firms acquire images, data that have a range of possible uses in fitness, business, civic planning

Arizona State University has been more of a tech hub than ever, with tricked-out cars cruising under the Tempe campus' University Bridge while young men nearby lug gadget-heavy backpacks past Old Main.

In separate projects, Uber and Google have been mapping a school that has become known for jet packs, robot swarms and NASA missions.

Uber’s work involves self-driving cars with sensors and gadgets on the roof of a small SUV that otherwise don’t give any indication of what makes them unique on Tempe roads. The cars have a human behind the wheel.

Google, meanwhile, is expanding its Street View feature, capturing close-up images of all of ASU’s campuses. The feature on Google Earth and Google Maps allows users to zoom in and see accurate depictions of buildings and sidewalks, anything a pedestrian could see walking down the street.      

Each relies on mapping technology that seems poised to expand.

“It is quite an amazing system,” said Stewart Fotheringham, a professor of computational spatial science in the School of Geographical Sciences and Urban Planning. Fotheringham teaches a class called Geographic Info Technologies (GIS 205), which introduces students to modern geographic information technologies, including cartography, GIS, remote sensing, global positioning systems and statistical analyses.

He explained how the technology works. 

“The video recorders Google people carry around with them record 360 degrees and are linked to the U.S. GPS satellite system, which consists of 24 active satellites about 12,000 miles above the Earth,” Fotheringham said. “Distances to four or more of these satellites give location on the Earth, plus elevation, so when you type an address into Google Earth, it looks up the coordinates of this location in a huge database and retrieves the imagery associated with this location.”

The imagery consists of satellite photographs and pictures taken from planes for Google Earth. Video recordings are taken by people on foot or bike and by cameras mounted to cars for Google Street View.

 

Andrew Ortiz is one of the field operators who troops around under the weight of a nearly 70-pound pack to get the shots the technology relies on. He and Daniel Quach took turns under the Google Street View Trekkers, a backpack outfitted with a camera system that’s used where vehicles can’t go.

It can be a tough task. “Our tolerance has been building,” he said. “I could probably walk about 2½ miles (before taking a break.)”

Google’s field operators worked around Tempe campus on March 6, starting in the middle of an area and walking out in concentric circles snapping pics with about a dozen cameras on a ball array at the top of their packs. In a mile the array will take about 500 to 600 shots, according to Ortiz.

Before coming to ASU, Google personnel worked with the university to come up with a priority list. They will shoot interiors in certain buildings, like ASU Gammage, but not every building on campus.

Uber cars have rolled past the Frank Lloyd Wright-designed Gammage and ASU’s School of Music, affectionately known as the “birthday cake building,” and the Ira A. Fulton Schools of Engineering countless times since their program introducing the vehicles to the Valley began in February.

Driverless cars need far more detailed digital maps than cameras and radar alone can provide, and Uber is recording every curb, stop sign and building to ensure their system works effectively. The cars include a pair of engineers in the front seats as safety drivers, according to a report in the tech-focused publication The Verge.  

Arizona Gov. Doug Ducey has hailed Uber’s self-driving pilot, releasing a statement saying his state’s participation “is paving the way for new technology and new businesses.”

The emerging mapping technology has a range of uses and applications.

“It's used by Realtors and house buyers, in in-car navigation systems, and by planners,” Fotheringham said.

Google, for example, takes a constellation of images that anyone could access from GPS satellites, then applies 360 camera technology and manages the the logistics of capturing video imagery on every road and street it can access, Fotheringham said.

Google’s data sets are turning out to be useful, as well, said Trisalyn Nelson, director of the School of Geographical Sciences and Urban Planning.

“Google Maps was developed originally so people could figure out where to go, and find their roots, and maybe check out an area before they arrive, but because these data sets are wall-to-wall — you have continuous data that maps a whole area with a lot of detail — we’re starting to see new, unintended uses,” Nelson said.

One example is fitness apps. Strava is a website and mobile app used to track athletic activity via GPS. If you’re biking, it tells you how far you rode, how quickly you got there and other similar information.

The people who run Strava realized they had their hands on a huge amount of data on where people were riding bikes. They repackaged it as Strava Metro and now sell the data to cities. More than 70 cities and organizations use Strava Metro to improve bicycle and pedestrian infrastructure.

“Getting data on where people ride their bikes in traditional forms is quite hard, because you have to stand at a corner to count people,” Nelson said. “Now we have this data that’s continuous through time, it’s continuous through space, and we can use it for urban planning to figure out where we should put in bike lanes. So this is a really good example of how new map technology starts off with one purpose — for tracking my fitness — and now we can use it to better understand cycling infrastructure.”

The expansion should be rapid, with tech and auto companies planning to share mapping data and trucking companies installing scanning systems. Next year, Volkswagen, which Uber uses in its self-driving vehicle pilot fleet, and BMW will install front-facing camera systems to detect obstacles. The systems will also passively map and send the data back. With millions of vehicles recording and relaying mapping data, there will be a geometric progression of data acquisition.

Unintended uses of mapping data will continue to balloon, Nelson said.

“We’re excited in our school about it, because on the one hand we do a lot of geographic information science, so we are the people who develop new technology to analyze map data,” she said. “On the other hand, we have all these people who are planning cities — the urban planning community — who are finding huge opportunities to use these new kinds of data sets to better understand what’s happening in a city, to inventory what’s there, to look at how it’s changing over time, to simulate future environments. Having access to these kinds of data opens up a lot of potential.”

It’s also opening up a new career field. A recent Forbes list of the top 10 growing jobs included two that fell in the category of working with maps and another technology. Computer science, statistics and geography are all taught at the School of Geographical Sciences and Urban Planning.

“We’ve seen this huge explosion in jobs for people who can work with maps, because maps are no longer something you draw,” Nelson said. “There are a lot of jobs associated with digital mapping technologies. In order to be really awesome, you need to combine it with something else. … These are very smart areas for ASU students to look for degrees.”

 

Top photo: As part of Google's mapping project, field operator Daniel Quach treks along the western part of the Tempe campus carrying the nearly 70-pound pack on March 6. He and Andrew Ortiz used the pack to help create pedestrian maps of all the ASU campuses last week. Photo by Charlie Leight/ASU Now

 
image title
Cars are like beer or ice cream: fine in moderation but bad to binge on.
March 13, 2017

ASU lecture features professor who says geographical information systems can help transportation

Sure, Google maps can get you where you’re going faster — but it can also help create a healthier, more sustainable city.

The rise of low-cost geosensors is changing the field of geography dramatically, and Ohio State University Professor Harvey Miller said it’s also “improving the scientific understanding of built environment characteristics that facilitate healthy and safe physical activity.”

Miller will give ASU’s School of Geographical Sciences and Urban Planning’s 16th Annual Malcolm Comeaux Lecture Tuesday evening at the College Avenue Commons on the Tempe campus.

With a focus on human mobility patterns, Miller hopes to put the wealth of data being collected through mobile phones, GPS systems and other devices to work toward more sustainable urban transportation.

“It’s a huge issue,” he said, “because humans have become an urban species. Over 50 percent of the world’s population now lives in cities. By the end of this century, that will be closer to 70-80 percent. You can’t have sustainable cities without sustainable transportation.”

ASU Professor Emeritus Malcom Comeaux, for whom the lecture is named, agreed that gleaning big data is making waves in geography.

“There are many jobs in the private and public sector that require GISgeographic information systems skills, and it has had a major impact on the discipline, in both research and teaching,” Comeaux said.

To find out how, ASU Now picked Miller’s brain on the subject in advance of his upcoming talk. Check out the Q&A below for his insights.

Editor’s note: Responses have been edited for length and clarity.

Harvey Miller

Question: How is geography changing?

Answer: Geography is going through a scientific revolution. It’s an ancient discipline, it goes back to the ancient Greeks. But for most of the history of geography as a field of investigation, data has been extremely hard to get. Back in the real old days, you’d send ships off in the ocean and wait a year for them to come back with a log and a map of where they’d been. People literally died for geographic information, it was so hard to extract from world.

Now, we’re drowning in data. Physical geography data, human movement data, data from mobile phones, GPS data, geosensor networksGeosensor networks monitor phenomena in geographic space. The sensors can be static or mobile, and can be used to passively collect information about the environment or to actively influence it., etc. You can put low-cost sensors on phones, bicycles or vehicles. You can collect an incredible amount of data about the world.

Also, we have much more powerful tools; computers have gotten much more powerful. So not only is data easier to get but our tools for analyzing data and simulating geographic phenomena — such as people moving through cities — have improved dramatically. It’s a really exciting time to be in geography. A whole new world is opening up to us.

Q: What can we use that data for?

A: A lot of people in cities — policy people, stakeholders — they want to make cities healthier places. They want to design environments that allow people to be as healthy as possible. But they need evidence.

There’s something called a “complete street,” where instead of a street just being for cars, it’s also for bikes and walking. Another design intervention cities are looking at is something called “active buildings,” where, for example, stairs are more prominent than elevators. They’re all great ideas, but they’re expensive and also sometimes controversial, so we need evidence that they work.

In the past, we just didn’t have strong evidence to show that these interventions had an effect because we could only record data at one moment in time and often relied on self-reporting. The problem with that is you can’t show change in behavior longitudinally over time and people tend to wildly exaggerate their physical activity levels.

Geodata is a different form of scientific evidence. Now, when something changes in the real world, in real time, we can track it and analyze how it happened. I just finished a project where I looked at a neighborhood in Salt Lake City, Utah, where I used to live. The city built a light rail line, and we got 536 people to wear GPS receivers to record their location in time and an accelerometer that records physical activity. They wore them for one week before the light rail was constructed, and then for one week after. We measured physical activity levels and time spent doing physical activity, and we found the light rail had a healthy impact.

Cars are like beer or ice cream. They’re fine in moderation, but you don’t want to binge on them.

Q: Have you learned anything surprising from this wealth of data?

A: People are complex. A lot of their behavior can be very counterintuitive. We’re finding a lot of surprising things. Travel is more complex than we assumed in the past.

One example is traffic counter data and air quality alerts. In Salt Lake City during the winter, they have really bad air quality so they use alerts to let people know. We wanted to see whether or not air quality alerts had an effect on travel. Were people actually driving less when the air quality was bad? We found the alerts actually increased people’s driving. The reason is an air quality alert system sends two messages. One is that the air is bad outside, so don’t run, don’t bike. But at same time, it says please be a good person and don’t drive. So people are faced with a dilemma.

Also, near the canyons, where people go to escape the bad air, traffic was worse because more people were trying to get out of the city. So there was a counterintuitive affect.

Q: How do you hope to see data like this affect cities of the future?

A: I hope it makes urban transportation more sustainable. That’s a huge issue because humans have become an urban species. Over 50 percent of the world’s population now lives in cities. By the end of this century, that will be closer to 70-80 percent. You can’t have sustainable cities without sustainable transportation.

Right now, the way our cities work, they’re automobile-dominated. Cars are like beer or ice cream. They’re fine in moderation, but you don’t want to binge on them. Cars are bad for cities, and they’re bad for safety. Somehow, we’ve come to accept a high rate of driving deaths, but we wouldn’t do that if it started happening with planes. We’d be up in arms.

Americans spend more on autos than they do on food and health care. We spend about $10,000 a year on automobiles. This is when the median income is around $40,000. That’s a big chunk of the budget for something that sits empty and stationary for 90 percent of its existence. It’s a real stress on the American household, and it makes driving very socially inequitable. There are better ways of mobility and accessiblitiy.

So I hope this data can help to reduce one of our biggest wastes of energy, biggest costs to health, biggest costs to our pocket books and something that’s changing global climate for the worse.

Q: Would you say services like Uber and Lyft are helping or hurting?

A: I live in central Columbus, Ohio. I drive as little as possible, but I do use Uber and Lyft because I can’t always bike or walk. But people are now buying cars to work for Uber and Lyft, which is another unintended, counterintuitive effect.

Columbus has a Smart City grant, so we’re going to have self-driving vehicles within four years. It’s all really exciting, but you can’t solve a car problem with more cars. You have to create transportation polyculture that offers many mobility options, including high-quality public transit. The most efficient way to move people is rail, but it’s expensive to build. We need to figure out how to build rail to support other forms of transit instead of competing with them.

There’s also the possibility of self-driving shuttles. We used to have something called jitneysA jitney was a small bus or car following a regular route along which it picked up and discharged passengers., but we regulated them out of existence. Mobility should be a service, not something that requires ownership. It doesn’t meant we can’t own cars, but we shouldn’t use cars to do every day mundane activities. They should be for recreation.

Q: What can the average citizen do to help?

A: Move into denser city centers or urban nodes. Live smaller and closer together. I know that runs against the grain of what some people want; some people want huge houses and big lawns, but that’s not sustainable. Suburbs are not sustainable. People have to sacrifice space and move into smaller, more dense, connected housing.

The other thing people can do is contact their local politicians and push for more local transit and biking lanes. We want to get people to bike. For a good part of the year, people should be able to bike. The No. 1 way to get people to do that is to build protective bike lanes. Research shows there’s a large amount of people who would bike if they didn’t have to do it in traffic. Another thing is walkability. A lot of city sidewalks are in bad condition and that does intimidate some people from walking. Those are the kind of design and policy things we need to ask our local city leaders to provide for us.