When Knightscope's security robots aren't , they actually have some serious work to do.
Silicon Valley-based Knightscope makes security robots that patrol areas like shopping malls to complement human security teams. The company's aim, according to CEO William Santana Li, is to give these teams "really smart eyes and ears to let them be able to do their jobs."
Knightscope added two more robots to its existing lineup that currently includes the K3 and outdoor K5, which made headlines forin a shopping mall.
The K1 is a stationary robot that can scan passers-by for weapons. Using millimeter wave technology, the K1 can sense the size and shape of objects. It's similar to the body scanning machines the TSA use except you don't need to walk into an enclosure. The company says it's suited for use at entry and exit points in airports or hospitals.
Then there's the 770-pound K7 that can traverse rugged terrain like gravel, dirt and sand on its four wheels. Rather than indoor patrols, this buggy-like bot is built for outdoor areas like wind farms or airfields. At the moment its top speed is capped at 3 mph.
Like the other robots, the new models have a similar suite of sensors including cameras, sonar and LIDAR. Using Wi-Fi, a cell connection or a combination of both, the robots process much of the data on board before sending it to a web interface accessible by the human security team.
"If you have an open air facility and three-o-clock in the morning, you have someone wandering around … let the guard then know and have the machine do the monotonous, computationally heavy stuff," said Li. "This is not intended to replace humans."
Knightscope provides the robots and the monitoring interface to companies at a rate of $7 per hour. The robots generally take a "coffee break" to recharge on docking stations once every 2.5 hours. During this time the bots are still fully operational, they just can't roam around autonomously.
One of the technologies that Knightscope is working on for future updates is what Li calls "audio event detection," which he likens to the opposite of voice detection. Instead of picking up individual voices, the robot will learn to differentiate sounds in the background, such as a footstep or glass breaking. That data could be used to localize the sound so a human security guard team can work out exactly where it came from.
The company is also working on an AI concierge, so you can talk to the robot and get responses while it's out and about.
It's Complicated: This is dating in the age of apps. Having fun yet?
Tech Enabled: CNET chronicles tech's role in providing new kinds of accessibility.