We've been hearing a lot about autonomous cars lately – and for good reasons. Driverless cars offer enormous opportunities for improved safety, convenience, and efficiency. Their proliferation may have as profound an impact on our society as conventional automobiles have had over the past century. But deploying autonomous cars widely is going to take a while, given the complexity of the application and the associated technological and regulatory challenges.
In the meantime, autonomous vehicles of other kinds are already on sale.
iRobot's Roomba vacuum cleaning robot is the most successful consumer robot of all time, with more than 14 million units sold. Of course, the Roomba has been autonomous from the beginning – the whole point of the product is that it cleans your floors while you're doing other things. But with their very basic suite of sensors, these vacuum cleaning robots have had limited capability to clean autonomously.
As Evan Ackerman writes in IEEE Spectrum: "Previous Roomba models can typically clean up to three rooms on a single charge; they can't handle more than that because, once they go back to the charging dock, they don't have a map to know where to resume from."
But in September, iRobot launched the Roomba 980, introducing intelligent visual navigation. According to Ackerman, "Roomba 980…can make its way from room to room, beeline back to its charging dock when its battery gets low, and then continue right where it left off. For the first time, you can rely on your Roomba to clean your entire single-level floor without supervision or assistance, which is incredibly awesome."
Another great example of vision-enabled autonomy comes from drone maker DJI. DJI has established itself as a leader in drones for consumer and professional photography and videography, and has shipped hundreds of thousands of drones. DJI's latest model, the Phantom 4, which went on sale last month for $1400, incorporates vision-based tracking and obstacle avoidance.
In its warehouses, Amazon has deployed 15,000 Kiva mobile robots to speed collection of items for customer orders. In hotels (a few hotels in the San Jose area, at least), robots from Savioke deliver snacks and sundries to guests.
So, while it may be a few years before you can use a self-driving car for your daily commute, today you can use an autonomous vehicle to keep your floors clean, shoot aerial video, bring a snack to your hotel room, or assemble your Amazon.com order.
I'm excited about the ways in which autonomous devices are already making our lives better. And it's clear there's much more to come. I believe that autonomous vehicles will proliferate rapidly in many applications, for two simple reasons: First, autonomous devices are often superior to manually operated ones in important ways, such as accuracy, safety, and efficiency. Second, key technologies, including computer vision algorithms and low-power, high-performance processors, have now matured to the point where it is practical to enable devices to be autonomous – including cost-sensitive and battery-powered devices.
Enabling autonomy is one of the most important uses of computer vision technology, and autonomous devices are becoming a big market for vision technology. Reflecting this importance, we've made autonomy a major theme at the Embedded Vision Summit conference (for which I organize the presentation program). The Summit takes place May 2-4 in Santa Clara, California. It's the only event in the world focused entirely on building better products using computer vision. This year's Summit includes several presentations on various aspects of enabling autonomy, including talks by visionary innovators like Jeff Dean of Google and Larry Matthies of NASA JPL.
If you're interested in learning about computer vision business opportunities, capabilities, technology, and techniques, I invite you to join me at the Embedded Vision Summit on May 2-4 in Santa Clara, California. This event, an educational forum for product creators interested in incorporating visual intelligence into electronic systems and software, is organized by the Embedded Vision Alliance. For details about this unique conference, and to register, please visit www.embedded-vision.com/summit.