Robots are precise machines — but the real world is anything but. A growing body of research is tackling one of robotics' thorniest problems: how to make robots act sensibly when they don't have complete information.
Yen-Ling Kuo is among the researchers leading that charge. According to IEEE Spectrum, Kuo — recognized as an award-winning figure in her field — has built her career around training robots to make educated guesses when facing uncertainty.
Kuo's path to robotics began in Taiwan, where, according to IEEE Spectrum, a childhood story about the physicist Michael Faraday sparked her lifelong curiosity about how the natural world works. She was also introduced early to Logo, a beginner-friendly computer program that used a turtle cursor to help children learn programming concepts — a small but formative on-ramp to the field she would eventually help define.
The ability for a robot to reason under uncertainty — to act confidently on incomplete or ambiguous information — is a foundational challenge in artificial intelligence and robotics. Most robots today still struggle when reality doesn't match their programming. Teaching them to make intelligent, probabilistic guesses rather than freezing or failing is key to deploying them reliably in homes, hospitals, and factories.
If researchers like Kuo succeed, it could mean robots that don't just follow rigid scripts but adapt — making them genuinely useful in the messy, unpredictable conditions of everyday life.