At the International Conference on Robotics and Automation in May, researchers from the Massachusetts Institute of Technology will present a decentralized planning algorithm for teams of robots. The technology factors in both stationary and moving obstacles.

The algorithm also requires significantly less communications bandwidth than existing decentralized algorithms, but preserves strong mathematical guarantees that the robots will avoid collisions.

With a typical decentralized group planning algorithm, each robot might broadcast environmental observations to its teammates, and all the robots would then execute the same planning algorithm, presumably on the basis of the same information.

The MIT approach maps out an obstacle-free region in its immediate environment and passes the map only to its nearest neighbors. When a robot receives a map from a neighbor, it calculates the intersection of that map with its own and passes the information on. The technique reduces robot communications.

The researchers are also testing a version of their algorithm on wheeled robots designed to collectively carry an object across a room of moving human beings. The experiment simulates an environment in which humans and robots work together.

Source

Also: Learn about NASA's Autonomous Guidance Algorithm for Auto-Pilot Spacecraft.