Practice makes perfect? You can bet the farmhouse on this! Computer scientist have started teaching robots the very human process of trial and error, in order for these machines in order to improve their movement.
Science Daily has the lowdown on this approach that is being applied to bots.
Computer scientists at the University of Leeds are taking first steps here, using artificial intelligence (AI) techniques to train a robot to find an object in a cluttered space like a warehouse shelf or in a fridge. The robot was then tasked to move it.
AI techniques of automated planning and reinforcement learning were put to good use here, with the ultimate aim of developing robotic autonomy to enable machines to assess unique circumstances presented in a task and find a solution.
This will allow the robot to transfer its skills and knowledge to a new problem.
Dr Matteo Leonetti, from the School of Computing:
“Artificial intelligence is good at enabling robots to reason – for example, we have seen robots involved in games of chess with grandmasters. But robots aren’t very good at what humans do very well: being highly mobile and dexterous. Those physical skills have been hardwired into the human brain, the result of evolution and the way we practice and practice and practice.”
Researchers used a robotic arm with an automated planning technology to see an apple surrounded by plastic bottles, cups, and other kitchen items.
It then calculated the moves required to reach it, on its own.
To address the problem of the arm knocking over other items in its path, the team employed reinforced learning AI that exposed the computer to around 10,000 trial and error scenarios to reach the apple.
And just as human do, the arm was able to learn which moves were more successful than other, and then adjusted its actions accordingly.
The team hopes that its findings could lead the way (pun always intended), towards enhancing the ability of robots that are working in warehouses to reach objects safely and carefully. It will also enable these machines in exercising greater care in their surroundings.