VP of Research at Samsung Talks About Their New Cooking Robot: Samsung Bot Chef (interview)
Sajid Sadi is the VP of Research at Samsung and is also the Head and Cofounder of the Think Tank Team at Samsung which is a group of people focused on solving complex technical and design challenges. He had also co-founded Tikatok.com; a platform that allows children to write and publish books and subsequently was acquired by Barnes & Noble within one year. Sajid was also a pioneer of the HTML5 revolution and helped to create one of the 1st browser-based editing platforms in the world which are creating software that is still used today to educate millions of children. He had also worked as a technology evangelist in the B&N Nook team to drive the development of recommendation-driven UI common to devices today. Sajid’s work in wearable technologies extends back to his time at MIT Media Lab where he had developed a wide range of wearable systems. Likewise, he had driven the development of the Samsung 3D 360 camera system, which enables users to step into any location instantly and in real-time.
His most recent work; the Samsung Chef Bot is pushing the boundaries of robotics. This invention brings robotics into the price range similar to home appliances and also advances deep learning technologies. His work helps to push forward what robotics can be to homes and small businesses today and in the future. At this time, he has filled over 70 patent disclosures across a variety of fields.
Can you tell us about the Samsung Bot Chef?
Bot Chef is Samsung’s next-generation robot. It is an AI-powered collaborative robotic arm that can use everyday kitchen tools. As a cooking assistant, Bot Chef can guide you through recipe steps while cutting, mixing, and seasoning as needed. There is no need to instruct this robot with complicated programming. Bot Chef can parse recipes and figure out which steps it can help out with. It can also understand the user’s voice commands to start things off and communicate with you.
Bot Chef also adapts to your particular kitchen, and if it can’t find an object that it needs, it will ask for your help. As soon as you place the item within view of the robot it will detect the object and figure out how best to grasp it.
Bot Chef also is designed with safety and affordability in mind. For this technology to really make an impact on our everyday lives, it needs to be affordable, and we need to feel safe around it. That is why we have engineered these robots from the ground up, using custom gearboxes and electronics that will help their price tag read like a kitchen appliance instead of a luxury car. The 3-finger gripper has been designed to pick up hard-to-grab objects like coffee pods but is also sensitive enough to know the difference between picking and squeezing. It is also safe: if it happens to get too close to you while it is holding something sharp, it either slows down or stops to keep your hands safe. In the case of impact, the robot’s mechanism is designed to merely bump into you, not bowl you over.
What other projects are being developed?
The Think Tank Team is an advanced product R&D lab, and as such, most of the work we do is kept under wraps. We do share our public projects online. The projects you see there are either product-path or close enough that Samsung chose to show the projects publicly.
How is Samsung advancing in the market?
(answering from the perspective of robotics)
Traditional robotics focuses on very specific, repetitive tasks. You see this all the time in a car factory: parts are held in very specific places, and the same operation is repeated by the same set of robots every time. This is not what a human being does at home. Every home is different, and home tasks change all the time in minute ways. Additionally, because these robots are “replaying” the same action over and over, they aren’t really aware of their environment. This makes them quite dangerous to be around. On a related note, the robots need to be programmed in very tedious ways, and this “programming” is very fixed and needs to be redone for the smallest change. We are changing all of that.
We are advancing the market in three critical ways:
Bot Chef is simple to use: you can talk to it to get it to do things. It understands your environment and adapts to the fact things in the home naturally move about. It is also able to download skills and do new things such as using new appliances and tools over time.
Safe: Bot Chef is designed from the ground up to work with humans. Advanced algorithms watch for obstacles, both human bodies, and objects, and plan motions around them when needed. If a body part is detected too close to a sharp or hot object, the system pauses to allow you to move away. And in the worst case, the hardware balances power against safety so that in case of a collision, it is no worse than bumping into a person.
Affordable: Robots currently cost about as much as a car, even for small machines. We have re-engineered every part of the robot to bring the costs down to the same order as kitchen appliances, making them available for the home.
Combined with our other AI and robotics efforts, we are changing the face of home robotics, and creating devices that care meaningfully for the individual and the family.
In what ways will AI and VR technology continue to grow?
As you can see from the work we demonstrated at CES, we believe that AI will come to the real world more and more. Right now, a lot of it is behind a screen, but we believe that AI will become seamless with our everyday life, whether it’s in the bedroom, kitchen, or car.
As for VR, unfortunately, I cannot comment on the topic.
What innovative technologies will still be growing in the coming years?
While I cannot, for the aforementioned reasons, talk about the specific areas we are investigating, I believe that the core of innovation is intuitiveness and a “sense of place” within the regular person’s life. Technology should fit into our lives seamlessly, and be transparent in its execution. Connections should happen instantly, everywhere. The cloud should be instantly accessible to extend the capabilities of devices. Robots should help do things for individuals. The list goes on and on.
One underlying theme there is communication, whether it’s machine-to-machine, machine-to-human, or human-to-human. I believe we will also see technology play a stronger role in us living a better and more fulfilling life. AI and automation will be big players here, but we will also need to innovate in sensors, algorithms, output devices, and even materials to pull our digital worlds closer to the physical one. These are just some general trends across the entire technology ecosystem.
What are three pieces of advice you would like to give to innovators?
This is always a tough question, but let me put a few big things out there not for doing innovation (which is to say, coming up with something cool in your head), but for taking innovation out into the world.
1. Make a difference. This is a complex one. First, you can only make a difference for people — companies and objects honestly couldn’t care less. Secondly, you have to be broad in your definition of difference. If you are an NGO, what difference you can make is not the same as if you are a tech company. But whatever you do, it will touch people. If you can touch those who are out of reach now or let people do something differently or better, you are making a difference. This means that every product can be considered from this perspective — and it’s the very core of whether your innovation will lead to a legacy or fizzle.
2. Execution is more important than anything else. I sincerely believe that there is nothing new under the sun. Someone somewhere has dreamt in their wildest dreams about what you are about to do. Stamp out any instance of not-invented-here syndrome and novelty-fetishism. If you believe you can execute it better, then make it happen — and create the common ground so that everyone works on it. Use what you can, and invent the rest. Remember, Facebook was not the first social network. Microsoft was not the first office software or OS vendor. Tesla didn’t make the first electric cars.
3. Be up-front on how your process works. Creating mysticism around innovation is the shortcut to failure. Innovation is everyone’s job because literally anything could give you the edge you need. Also, you don’t have an oracle to show the path — innovation takes a lot of trial and error. Safeguard failures, because everything until the final product must be one. Chronicle the stories of accidental success so people take off their blinders and look at every result critically. Most of all, regardless of what you tell the world, make sure to keep the real story alive inside the team so that your team doesn’t confuse its marketing story for its process.
