New Algorithm enables Robot to Learn through Trial and Error UC Berkeley's BRETT (Berkeley Robot for the Elimination of Tedious Tasks) is capable of learning through trial and error, like humans. New algorithms developed by researchers empower the robot to master tasks through trial and error, ridding the need of pre-programming. Among many tasks, it can perform is assembling a toy, and the best thing is it keeps trying figuring out the way to accomplish the task until it finally done.

The researchers are optimistic that the further development of the robotic technology will enable robots to handle lots of data. The technology gives rise to artificial intelligence to allow robots to do anything their designs allow. For example, building something or playing a new sport.

New algorithms developed by researchers from UC Berkeley brought this trial and error process to robots. UC Berkeley said in a press release that the technology is a giant leap in the field of artificial intelligence.

The technology enables the robot to perform tasks like putting a clothes hanger on a rack without feeding details into it about its surrounding.

�The key is that when a robot is faced with something new, we won�t have to reprogram it. The exact same software, which encodes how the robot can learn, was used to allow the robot to learn all the different tasks we gave it�, said UC Berkeley Professor Pieter Abbeel.

The new technique will be presented by Abbeel and his fellow researchers at the International Conference on Robotics and Automation in Seattle on May 28.

According to Trevor Darrell, director of the Berkeley Vision and Learning Center, it is very essential to empower robots to learn and acclimatize to their surroundings before they are declared fit for use in homes. The new technique is inspired from the neural circuitry in human brain.