Anki's Vector Is a Little AI-Powered Robot Now on Kickstarter for $200

Is this the personal home robot we've been waiting for? No, but Vector will make you smile

3 min read
Anki Vector robot
Hey Vector, what's the weather in Seattle?
Photo: Anki

San Francisco-based robotics company Anki is launching a new robot called Vector today. It drives on little tank treads, has an expressive LCD face, and is constantly moving around, making beeps and bloops. Vector resembles Anki’s previous robot, Cozmo, but is packed with more technology—and character.

Anki says Vector is designed as a robot companion and helper for people at home. But don’t expect much: The robot has a tiny voice and can barely push a coffee mug around. Its little arms can lift only a special plastic block. What Vector is good at, Anki says, is engaging with people.

The robot can certainly get your attention. At one point during my interview with Anki’s chief strategy officer, Patrick De Neale, Vector interrupted us to demand a fist bump. De Neale promptly complied.

Vector is available starting today on Kickstarter at a discounted price of US $200 (it will hit stores on 12 October with a $250 price tag). Kickstarter backers will also get early access to the software development kit (SDK), which will be available to all users only later this year.

Vector is powered by a 1.2-GHz quad-core Qualcomm Snapdragon and has a wide-angle HD camera, a single-point laser for mapping and navigation, a four-mic array, and capacitive touch sensors. And while its speech capabilities rely on the cloud, other functions, like detecting if there are people nearby, use a convolutional neural network running on its onboard processor.

Vector is Anki’s fourth product. First came the robotic racing cars Drive and Overdrive. Two years ago, the company unveiled Cozmo, which became a best-selling toy on Amazon. But Anki says Vector is not a toy. The company is apparently targeting a group of tech-minded grown-ups who have always dreamed of having R2-D2 in their living rooms (ahem, we know some of them).

Anki vector with charger and cube Image: Anki

Of course, the idea of a personal robot is something companies have long been trying to bring to market, but success has proved elusive. From Topo in the 1980s to, more recently, Pepper, Jibo, and Kuri, personal home robots have struggled to strike a balance between cuteness, usefulness, and cost. So we find a bit worrisome that Anki is raising people’s expectations by repeating some of the same claims we’ve heard before, including that Vector is going to become “part of the family.” (Noooooooo!)

Anki, which has raised over $200 million from investors like JP Morgan and Andreessen Horowitz, argues that there’s a big difference in its approach. Kuri and Jibo cost around $700, and Pepper even more. Anki wants to be able to produce and ship high volumes of robots with lower prices. The company says it has sold more than 1.5 million robots and had nearly $100 million in revenue last year.

During our demo, De Neale showed Vector doing things like setting a timer, checking the weather, converting units, building a map of its surroundings, and learning to recognize a face. I asked if those things—some of which users may prefer to do with their phones or Amazon Echos—will convince people they need a robot like Vector at home.

De Neale explained that Anki knows it can’t build Rosie the Robot today. So the company’s plan is to build up its technology, iterating and shipping ever more powerful robots. He noted that while Cozmo has 340 parts, Vector has about 700. And while Cozmo has to be tethered to a smartphone to operate, Vector uses its Snapdragon processor and a Wi-Fi connection to stay always on.

Anki Cozmo vs. Anki Vector Image: Anki

Indeed, I can see the potential in the robot. Vector will get better over time, as Anki pushes updates to its software. And then there’s the SDK, which will allow users to build new capabilities and maybe even create apps that they can share.

Still, I pressed De Neale on why someone would want to have Vector today.

“It comes around to creating the notion of relationship, to creating positive engagement with you on a regular basis in a way that is not demanding, that is not annoying, in a way that you sense value in it,” De Neale said.

“And if we do our job right,” he added, “you’ll find yourself kind of feeling a level of empathy for this product that, at the end of the day, is a collection of electronics and plastics and some motors.”

[ Anki ]

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

“I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

This article is part of our special report on AI, “The Great AI Reckoning.”

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less