Issue 103, October 2011
Meet Nash – NTU Advanced Smart Humanoid
NTU unveils Singapore’s first human-sized smart robot

Much like the famous golden protocol robot from Star Wars, C3PO, Nash (NTU Advanced Smart Humanoid) can observe its environment, gesture, walk, climb stairs and follow verbal instructions, making it possibly the smartest humanoid robot in the world. It has a mind capable of understanding meaning in words, and stands “a-head” at 1.8m, looking similar to robots found in Hollywood films, like the one in i, Robot, starring Will Smith.

Not only is it tall and smart, this 80kg “dreambot” is sturdy as well, resisting pushes and pulls from different directions, thanks to force sensors installed in its ankles; an impressive feature as most humanoid robots can barely stand on their own.

Birth of Nash
Its inventor, Professor Xie Ming, from the School of Mechanical and Aerospace Engineering, hopes that Nash will soon be able to interact with people and help them with menial tasks, such as carrying and lifting objects, and assisting the elderly.

The project to build a human-sized robot started from 2006, when Prof Xie set out to do what no one else in the world had done – to integrate human-like locomotion, manipulation and intelligence into a low cost human-sized platform.

Prof Xie, who is also the editor-in-chief of the International Journal of Humanoid Robotics and one of the world’s leading authorities for humanoid robots, already had ideas on how to overcome the limitations that prevented scientists worldwide from designing mentally and physically capable robots back in 2003 when he wrote his book, “Fundamentals of Robotics”.

“It took us over two years to come up with the design of a two-legged humanoid robot which could mimic the hand, arm and leg movements of a human body,” said the 48-year-old scientist, who has a start-up company dealing in robotics technology.

Prof Xie explained: “That wasn’t the hardest part, as the biggest challenge was to make the robot walk on uneven terrains without falling, and to recognise push and pull forces so it can resist and counterbalance itself. We have been trying very hard to solve these problems through the use of specific planning and control algorithms, which took us the better part of the last three years.

“Now, we are in a better position to make this a social robot, one that can talk and respond to humans, to understand us and to do what we ask of it to do. This will be a future challenge because it relies on cognitive linguistics; to make the robot understand what we are asking it to do, with the use of human languages instead of numerical values used in computer programming.”

This robotics project is supported by and built in NTU, with various parts of its technology developed and adapted from other related research projects.

© Alumni Affairs Office