There is a broad range of applications for mobile robots exploring unknown environments, including planetary exploration and support of first responders after a disaster. Our research focuses on missions with multiple robots acting in a team; teams can explore faster than single robots but in turn require coordination and information exchange and fusion. Solutions for online path planning, decentralized coordination, wireless communication, and energy efficiency are being developed.

For conducting practical experiments, Torsten Andre and students developed an integrated software solution in the Robot Operating System (ROS) for coordinated multi-robot exploration. It allows teams of robots to communicate over ad-hoc networks to coordinate exploration and merge mapped areas. Andre’s work furthermore focuses on the collaboration of robots, e.g. clearing blocked paths during exploration. He developed a heuristic that enables robots to decide when to collaborate and whom to collaborate with. This decision is not only based on the probability that an alternative path can be taken to bypass the obstacle but also on the topology of the robot network.

Another aspect of multi-robot exploration are endurance missions where energy efficiency is the focus. Micha Rappaport developed an energy-aware exploration strategy that allows robots to adaptively decide where to explore and when to recharge. The robots shall then collaborate to further push the energetic horizon. “The idea is that a team of robots can explore areas beyond the reach of an individual robot,” Rappaport explains. This is achieved by forming teams where one robot is exploring further and further away while others assist as mobile rechargers.