Robots that can work together as a team to serve thirsty patrons may soon replace human bartenders and waiters.
A team of engineers has designed robots that can work together to pour and serve beers to humans. Researchers from Massachusetts Institute of Technology (MIT)'s Computer Science and Artificial Intelligence Laboratory (CSAIL) were able to come up with a system of robots that can work together in a bar setting.
The system is consists of one PR2 robot, which acts as the bartender cracking open beers, and two other four-wheeled "Turtlebots" that act as waiters taking orders and delivering the suds.
Robots that do human tasks are not new. Machines that mix drinks with the help of an app and robot bartenders that are capable of creating cocktails exist but the notable thing about these bartending robots is not their individual skills but their teamwork.
Among the greatest challenges of getting robots to work together is the uncertainty involved in the human world given their limitations as machines. Communication and sensor problems, for instance, could pose problems in the system.
"Each robot's sensors get less-than-perfect information about the location and status of both themselves and the things around them," said MIT graduate student Ariel Anders. "On top of that, robots often are not able to communicate with one another, either because of communication noise or because they are out of range."
Anders said that these limitations prompted them to come up with algorithms that would make it possible for the robots to engage in higher-level reasoning about their behavior, status and location. The machines were able to manage uncertainly as they work together with sophisticated communication system and more hands-off programmatic approach.
Bartending is a good way to demonstrate the teamwork of a three-robot system but the team sees application for their system beyond bartending.
The system may have applications with more crucial implications in the real world. The researchers want to come up with robots that can help in unpredictable and high pressure settings such as those seen in search and rescue operations and hospitals.
"We introduced an extended MacDec-POMDP model for representing cooperative multi-robot systems under uncertainty using a high-level problem description. We also developed MDHS, a new MacDec-POMDP planning algorithm that searches over policies represented as finite-state controllers," the team wrote in the journal Robotics Proceedings. "In the bartender and waiters problem, MDHS was able to automatically generate controllers for a heterogenous robot team that collectively maximized team utility, using only a high-level model of the task."
