Gartner Blog Network

CES 2019 Day Two – The Rise of Reinforcement Learning Robotics

by Werner Goertz  |  January 10, 2019  |  2 Comments

In my previous blog (see: I argued that value creation from hardware is becoming subordinated to that created by ecosystems and AI. Nowhere is this more manifest than in personal robotics: for many years, CES has shown plenty of toy robots, many for children’s education and entertainment. The expected democratization of reinforcement learning will open new value creation in retail, hospitality, logistics, and of course, personal use cases.

Samsung’s suite of mechanical helpers is not an yet an example of adaptive reinforcement learning AI, but its concept announcement is indicative of what to expect:

– Bot Care is an entry point into robotics-enabled personal care systems that can be extended into remote healthcare and elder care ecosystems

– Bot Retail offers an additional interactive service channel for enterprises such as restaurants, some of which are exploring omni-channel UIs and delivery modes (see Panera Breads, for example)

The challenge for reinforcement AI-enabled robotic devices is the intersection of the digital and the analog domains. While rapid and disruptive progress can be expected from the evolution and democratization of AI in the digital domain, the laws of physics in the real world are more difficult to overcome:

– Gravity is (literally) holding us down! It causes us to devise complicated mechanical apparatus to lift, accelerate and manipulate objects.

– Energy storage is not evolving like Moore’s law is: the energy density of Li-Ion cells improves at only 3-5% YoY.

– Stuff breaks! In the digital domain, malfunctions are fixed with a reset button or an OTA software update. In the real world, a broken belt, gear or motor requires mechanical intervention.

I expect reinforcement AI enabled robotics to be a major personal technologies trend by the end of 2019 and beyond. Solutions presented at CES this year are still programmed in the traditional sense, but by CES 2020, we will see more robots that learn and adapt to their environment and become acquainted with the physical, sensory world and the humans they are designed to assist.

Additional Resources

View Free, Relevant Gartner Research

Gartner's research helps you cut through the complexity and deliver the knowledge you need to make the right decisions quickly, and with confidence.

Read Free Gartner Research


Werner Goertz
Research Director
1 years at Gartner
21 years IT Industry

Werner Goertz is a Research Director within Gartner's Personal Technnologies team, where he covers personal devices (smartphones, PCs/Ultrabooks, tablets/ultramobiles and wearables) and IoT. A special emphasis of his research lies in the Human Machine Interface (HMI) and multimodal I/O technologies: voice/speech processing and recognition, facial recognition and eye tracking, biometrics and motion/gesture control. Read Full Bio

Thoughts on CES 2019 Day Two – The Rise of Reinforcement Learning Robotics

  1. Mark Cundy says:

    Hi Werner, great observations, particularly how you tie together today’s observation of Reinforcement AI enabled Robotics with last month’s predictions (12/21 blog). Your view that the symbiotic yet so-far-unbalanced level of innovation in the analog vs. digital realms of technology is enticing. I was unable to attend CES this year, but am curious as to whether you have observed anything this week that confirms your prior prediction that we’ll see unprecedented levels of technology disruption in the next 3 years? Certainly eminent futurists (Kaku, Kurzweil, and others) have longer term views, but what about near term?

    Where in the analog realms of technology should astute observers look for breakthroughs to happen?

    • Werner Goertz says:

      Thanks, Mark. I am very familiar with the writings of Ray Kurzweil and Michio Kaku. Sadly, I have no magic innovation bullet when it comes to analog. In software, when you hit a bug you do a reset. Vendors fix the bug with over the air updates. In analog, however, when a belt snaps or a robotic motor fails, you’re SOL. The analog domain will slow down robotic mass adoption. Check and see if you can find something…

Leave a Reply

Your email address will not be published. Required fields are marked *

Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.