Automated Planners for Better Strategies

Editorials News | Feb-20-2017

Automated Planners for Better Strategies

Researchers from MIT's Computer Science and Artificial Intelligence Laboratory are trying to improve automated planners by giving them the benefit of human instincts. The scientist have added this intuition while considering that even the best planning algorithms still aren't as effective as human beings.

The scientists have created a solution by encoding the strategies of high-performing human planners in a machine-readable form. After programming the new device the scientists analyzed an increase in the performance of competition-winning planning algorithms by 10 to 15 percent on a challenging set of problems. The researchers are presenting their results this week at the Association for the Advancement of Artificial Intelligence's annual conference.

Every other year, the International Conference on Automated Planning and Scheduling hosts a competition. During this event, computer systems designed by conference participants try to find the best solution to a planning problem. The planning problems include scheduling flights or coordinating tasks for teams of independent satellites.