No way an Alexa on wheels, Astro, the first domestic robot assistant by Amazon has been provided with an unique persona that’s all its own. From adding eyes to the display, to a whole host of sounds, the device really comes alive and moving.
Astro will bring all of Alexa features like music, news, podcasts, and timers directly to their users. If doing a video call, but will move with its owner around the house, so that he/she can continue the conversation.
Like Alexa, which has continued to get smarter since it was launched seven years ago, Astro will get smarter and more capable over time too.
Astro will be available for an introductory price of $999.
From 28th September people interested in acquiring the robot can sign up to request an invite. Amazon plans to start granting invitations and shipping the device later this year.
There are no fixed roads at home, the first problem to be solved for a moving bot
How can a tech assistant on wheels move around at home without bumping against the school bag or the toys children have left scattered all over the floor?. There are no fixed roads at home and this was one of the first problems to solve for Astro to be actually convenient and operative.
That’s why Astro features embeded software and tools for autonomous movement, home mapping, visual understanding, and more specially developped by Amazon. The company has leveraged AI in many new ways, including using deep neural learning to map anchor points in the home, and building dynamic new SLAM (simultaneous localization and mapping) algorithms that are constantly refreshing.
Privacy protection the second big problem to be solved
In addition to many of the privacy features customers are used to with Echo devices, Amazon built brand new features for Astro:
Users can set out-of-bounds zones to denote areas where the owner don’t want Astro to go.
Turning on “do not disturb” the user can minimize how much Astro moves during certain times of the day or night.
Amazon has also embedded AI processors inside of Astro to allow for as much edge computing as possible.
