A new system designed at Georgia Institute of Technology can control a swarm of robots, simply by using one finger to shift a beam of light. Each robot is also flexible enough to develop its own plan, based on input received from others in the swarm.
“It’s not possible for a person to control a thousand or a million robots by individually programming each one where to go,” said Magnus Egerstedt, Schlumberger Professor in Georgia Tech’s School of Electrical and Computer Engineering, in a press release. “Instead, the operator controls an area that needs to be explored. Then the robots work together to determine the best ways to accomplish the job.”
Some aspects of the system are relatively straightforward: moving a finger around on the screen moves a light projected on the floor. As the light moves, so does the swarm. What makes this system unique is that the robots are in constant communication with each other regarding their location and the amount of light sensed in their immediate area. This allows the robots to work as a team. For example, if the operator puts two fingers in different locations on the tablet, the machines will split into groups to cover both areas. Or if one robot is sensing a lot more light than another, and the objective is for all to sense an equal amount of light, then the robot in the bright spot will move out of the way so that another can taken in some light. All this is possible because each robots can change course autonomously at any time, based on the information it is continuously receiving from the other robots in the swarm.
“The field of swarm robotics gets difficult when you expect teams of robots to be as dynamic and adaptive as humans,” Egerstedt explained. “People can quickly adapt to changing circumstances, make new decisions and act. Robots typically can’t. It’s hard for them to talk and form plans when everything is changing around them.”
The researchers involved in the project say the algorithm used to control the swarm could be relevant in several fields that involve large numbers of robots working together, such as manufacturing, agriculture, and rescue operations.