Simulation of Multi-Robot Teams with Flexible Level of Detail

A key methodology for the development of autonomous robots is testing using simulated robot motion and sensing systems. An important issue when simulating teams of heterogeneous autonomous robots is performance versus accuracy. In this paper the multi-robot-simulation framework (MuRoSimF) is presented which allows the flexible and transparent exchange and combination of the algorithms used for the simulation of motion and sensing systems of each individual robot in a scenario with individual level of realism. It has already been used successfully for the simulation of several types of legged and wheeled robots equipped with cameras and laser scanners. In this paper the core functionalities of MuRoSimF are presented. Existing algorithms for simulation of the robots’ motions are revised. Newly added features including the execution of the simulation on multi core CPUs and two different algorithms for the simulation of laser scanners are presented. The performance of these features is tested in an urban scenario using wheeled robots.