Download the authoritative guide: Cloud Computing 2018: Using the Cloud to Transform Your Business
SANTA CLARA, Calif. - Mapping is one of our oldest activities and remains key to many recent innovations and those still to come. A panel of experts here recently discussed advances in geospatial technology and its connection to such things as driverless cars and even augmented reality.
Brian MClendon, former head of Google Maps and now Vice President of Advanced Technologies at Uber, says a key to truly driverless or autonomous cars will be better maps. But that’s only the beginning. Google’s own self-driving cars, for example, have sensors that collect all sorts of data on road conditions, other vehicles on the street, pedestrians, cyclists and so on. As fleets of these cars emerge and the big data they’re collecting gets ever bigger, it promises to improve the safety and effectiveness of self-driving cars.
“We can use that data for deep learning so the cars can better identify pedestrians and what other cars are doing and relay that all off to a deep network that runs live on the car,” said McClendon, during the panel sponsored by the Churchill Club.
That’s the theory. In practice he says that getting that much data to the cloud is going to be challenging if it needs to be processed in real time. He adds that there will also have to be decisions or a system to determine how much data is relevant and how much should be discarded or sent to archival storage if it can’t help drivers immediately.
Meanwhile, human drivers will continue to benefit from technology advances designed to improve safety. For example, augmented reality could help drivers navigate the roadways and provide richer information of surroundings. Augmented reality “augments” things in the real world versus virtual reality, typically deployed via glasses or headgear, that puts the user in a whole new “virtual” world.
After the panel, McClendon told Datamation he sees the potential for augmented reality that’s built in to the windshield offering drivers such aids as a street map of what’s ahead or information on historical buildings as the car approaches them. But these systems will have to be very carefully designed from a safety perspective to not interfere or distract the driver said another panelist Jane Macfarlane, chief scientist and head of research at mapping and navigation company HERE. “If I’m going to interfere with your driving process, you have to be very careful about how you deliver it,” said Macfarlane. “In the short term it’s going to be a challenge to keep it safe.”
On the other hand, in a driverless car, augmented reality might prove to be a useful guide to the passenger to see landmarks in a new city and providing information.
Drones Help with Mapping and Construction
Drones are also being used to help with navigation, including at construction sites. Panelist Christian Sands, CEO and founder Skycatch, a company that makes software for drones, says customers have used drones on big construction sites like the San Francisco 49ers’ Levi Stadium to map out what needs to be done.
It’s a fast-changing industry. Sands says drones are helping companies gather real-time data that is by definition more current than printed and most online maps. “We’ve been working with clients where we have given them new data every day and now they’re saying that’s not fast enough,” says Sands.
Even collecting visual information from a drone requires processing on the ground, but Skycatch aims to shorten that turnaround time. “We’re building technology to do the data processing on the UAV (unmanned aerial vehicle) itself so it happens automatically and ties into the workflow,” said Sands.
He also expects rapid development of machine-to-machine activity where, for example, a drone would help identify something that needs to be moved at a construction site and direct a bulldozer to do it. “Five-to-10 years from now a lot of these activities are going to be automated,” said Sands.