Wednesday, September 23, 2015

Congrats KAIST friends!

Was 2007 when I had the pleasure and honor to meet Jun Ho Oh, professor of mechanical engineering at the Korea Advanced Institute of Science and Technology, who led the KAIST Team, who won in June 2015 the worldwide recognized DARPA Robotics Challenge (DRC) and its first prize of 1 Million Dollars.

Professor Ho Oh, came on 2007 from South Korea to Campus Party, the biggest LAN event in the world, were I was leading the organization of the Robotics Area (CampusBot), with his team from KAIST and Albert Hubo Robot. In that moment Hubo had a fully motorized Einstein face, able to reproduce not only mouth movements but face expressions. 

I found in Professor Ho Oh not only a brilliant mind now worldwide recognized, but also a very nice, sincere and friendly person, that was managing his team with a very close implication and understanding. Their robot did what was expected. In the picture I’m with him and with our respective robots: Hubo and Sam.


Professor Ho Oh and his team continued working in Hubo robot and improved it in order to perform real tasks in a real world. Therefore they won the prize. They deserved it. 

DARPA organized the DRC after the Fukushima disaster in Japan, expecting to advance the field of disaster robotics. The DRC Finals called for 23 teams of semi-autonomous robots partly controlled by human operators. The robots created by universities and companies included legged robots, wheeled robots, and hybrids robots as well. The following image shows the DRC Finals team standings, being on first position Hubo of TEAM KAIST from South Korea, on second position Running Man (Atlas) of Team IHMC Robotics from the Institute of Human and Machine Cognition in Florida/USA and on third position Chimp of Tartan Rescue Team from Carnegie Mellon University in Pennsylvania/USA.


The Robots had to perform things like drive a car, open a door, drill a hole in a wall, walk up stairs, open a valve and tread uneven ground. Also the Internet network that teams used to communicate with their robots at times didn’t work for 30 seconds or more, to simulate real-life conditions. Therefore this was a very difficult set of tasks to complete. Most of the robots didn’t perform some of the tasks, even most of them fell down sometime during the execution. But from my point of view it was a very interesting competition. People criticize that most of the robots failed, but most of the people don’t understand how difficult is to make robots perform like humans in real world. 

We made a world for humans. Everything in this world is prepared for humans (stairs, chairs, doors, vehicles), and this is the reason why most of the robots at DRC where humanoid robots, just to ease the execution of the tasks, but really is not a question of creating robots exactly like humans. There are better options, and teams like KAIST understood it when they decided to give HUBO the possibility of moving using legs or wheels, as well as being able to rotate its torso 180 degree. This couple of things between others showed that KAIST team did something great: Thinking very creatively, trying to find ways to even improve the human body limitations in some ways. My congratulations to the KAIST team. I wish them the best!

Recommended Links: