Life in the Robotics Lab

Davide Faconti was interviewed by Sander Olson from NextBigFuture!

Here you have the entire interview....

Sander: The PAL project is focused on humanoid robotics. What is the current status of this project?

Davide: The PAL project began about six years ago with 4 workers. Today there are about 25 researchers, and we are focused on rapidly improving these humanoid robots, in particular the REEM robot. We are headquartered in Spain but are owned by investors in the United Arab Emirates.

Sander: What are the main differences between the REEM-H and REEM-B robots?

Davide: The REEM-B robot is a prototype of a humanoid robot, and won't be commercially sold. The REEM-H2 robot is designed for the                                           REEM-B at the labcommercial market. It is wheeled and interaction occurs via a touch-screen on the chest.

Sander: Wouldn't voice commands be simpler?

Davide: The technology of voice commands has yet to be perfected. Voice commands will eventually be ubiquitous on robots but currently the technology is not sufficiently reliable, accurate, or robust to use.

Sander: Can REEM-H2 recharge itself?

Davide: Not yet. But the battery can power the robot for 8 to 10 hours. But we've designed the H2's to have batteries that can quickly and easily be replaced, so we don't see this as a problem.

Sander: What tasks can the H2 currently accomplish?

Davide: It can lift objects up to 8 kilograms, so it can transport small packages. It can move up and down ramps, and can navigate indoors using laser scanners. It can be used for video-conferencing, web access, and it can recognize faces and voices.

Sander: What tasks can REEM-B accomplish?

Davide: REEM-B was introduced 2 years ago and is targeted primarily at research labs. It is capable of face recognition, it can lift objects up to 14 kilograms, and can walk. It cannot yet climb stairs reliably, or operate outdoors.

Sander: How does REEM-B compare to Asimo?

Davide: The Asimo is a beautiful piece of engineering. But Honda has been developing Asimo for almost a quarter century, so they have a lengthy head start. REEM-B can lift heavier objects than Asimo, but unlike Asimo it can't run or dance.

Sander: Is there a REEM-C under development?

Davide: Yes, the target is to release REEM-C sometime in 2011. It will be more robust, it will have a greater range of motion, will be able to climb stairs, and will operate up to three times as fast as REEM-B. We also want to reduce the manufacturing costs. REEM-C will have 46 motors, with 16 motors just on the hands, and should be able to operate for 2-3 hours on a battery charge.

Sander: Will REEM-C be targeted to consumers?

Davide: No, it will be sold primarily to research labs. Humanoid robots are currently too expensive, difficult to operate, and clumsy to be useful to consumers.

Sander: Is software currently the main bottleneck for robotics?

Davide: I think it is. The software is not up the tasks we need it to do, which include reliably responding to voice commands, navigating in the real world, and interacting with objects. The good news is that there are many groups worldwide in the process of improving software programs that can make it possible to perform these tasks. Hardware will not be our major bottleneck once we get to mass production and costs are reduced.

Tags: Davide Faconti, PAL Robotics, REEM-B, REEM-C, REEM-H2

The design of our robot REEM: a well studied topic

Hi, I am Jan, and I am the Business Developer here in PAL Robotics. A very challenging job, in a company loaded with very intelligent and ambitious engineers, and in a world where we are at the very early stage of introducing robotics in society.   I just want to share a couple of words with you on one of the main issues of our robots: its design.

My task here is to sell these cute robots we make. But…how do you sell a robot? Many references from the market I did not get – there are for the moment not too much robots around, which, in a certain way, is good for us!  So we did a careful study to check what´s been done, and what failed and why and so on…and actually quite soon you find out that all these smart engineers worldwide make wonderful pieces of technology, but quite often they lack a nice design.

Why do so many people buy iPods and iPhones (see www.apple.com)? Because it is such a sleaky, slim and sexy tool! Yes, definitely in the case of Apple´s bestseller, the design is not the only, but a very important sales argument. And what counts for Apple, counts for a lot of brands and products.

So, why do robots not sell like iPods? ;) Well, there are of course hundreds of reasons (lack of offer, price, security, applications,…) , but one fact is surely people do not feel comfortable yet with robots, a.o. because the physical outlooks of the robots are not very attractive, and because they do not invite people to interact with the robot.

We spend a lot of time and effort in looking for a very aesthetical outlook for REEM, our latest prototype. Just to give you an idea; we organized public demonstrations (although with another model) to check out the reactions of the public (you can see some videos of these on our website, we paid attention to the movements of the robot and how that influences the interaction with people and we gave it a soft voice which is nice to listen to.

The result you will be able to see on both our blog and website – I hope you like it!

Please note that the design of REEM has been developed in-house, contrary to the design of  REEM-C, the new biped robot, for which we have a contest running on our blog.

Tags: Apple, humanoid, PAL, PAL Robotics, Podcast, REEM, REEM-C, robot, service robot