Date:12 May 2014
I’ve encountered many robots over the years, from relentlessly boring (and slightly intimidating) industrial machines that weld car bodies with monotonous precision to disconcertingly anthropomorphic characters that mimic human emotions, dance to rock music, ride bicycles, kick soccer balls and keep elderly people company when they feel neglected.
I like them, and not only because I grew up reading Isaac Asimov’s books. I recall a conversation with a charming Japanese man in Barcelona many years ago – he was the genius behind the “face” of Honda’s ASIMO robot – about the so-called “uncanny valley”, defined by Wikipedia as “a hypothesis in the field of human aesthetics which holds that when human features look and move almost, but not exactly, like natural human beings, it causes a response of revulsion among some human observers”.
He was emphatic that a robot should not attempt to replicate human features, but rather hint at them – that is, with a vaguely humanoid head, perhaps a vestigial mouth, and a more or less neutral “expression”. However, clever and thoughtful people like Cynthia Breazeal disagree. At MIT (the Massachusetts Institute of Technology), she and her team are building robots with social intelligence that communicate and learn in the same way that people do.
Will Breazeal’s work, and similar research being conducted in labs all over the world, change the way we live in the future, and the way in which we interact with intelligent machines? Short answer: yes, without a doubt. Is this a good thing? Absolutely. and if it means your rivals on the baseball field will grin at you in a disconcertingly machine-like way, then so be it. There are already robotic soccer tournaments, so a robot baseball player is entirely possible in the near future. Scientists have even developed a sensitive “skin” for robots, for heaven’s sake.
Meanwhile, in a new breakthrough, researchers at the EPFL (École Polytechnique Fédérale de Lausanne) in France have built a robot that’s capable of reacting without human intervention and grasping objects with complex shapes and trajectories in less than five hundredths of a second. With its palm open, the robot is completely motionless. A split second later, it suddenly unwinds and catches all sorts of flying objects thrown in its direction – a tennis racket, a baseball, a bottle.
The arm measures about 1,5 metres long and keeps an upright position. It has three joints and a sophisticated hand with four fingers. It was programmed at the Learning Algorithms and Systems Laboratory at EPFL and designed to test robotic solutions for capturing moving objects. It is unique and rather spectacular because it has the ability to catch projectiles of various irregular shapes in less than five hundredths of a second.
Says Aude Billard, head of the lab: “Increasingly present in our daily lives and used to perform various tasks, robots will be able to either catch or dodge complex objects in full-motion. Not only do we need machines able to react on the spot, but we also to predict the moving object’s dynamics and generate a movement in the opposite direction.”
The ability to catch flying things (once again, think robot baseball player) requires the integration of several parameters and reacting to unforeseen events in record time. “Today’s machines are often pre-programmed and cannot quickly assimilate data changes,” adds Billard. Consequently, their only choice is to recalculate the trajectories, which requires too much time from them in situations in which every fraction of a second can be decisive.”