Data fusion and software integration in robotics

Intelligent robots are systems capable to operate autonomously and to solve complex tasks, which desire sensing and fast situation recognition, intelligent decisions, and subsequently an efficient physical action. They process on-line a vast volume of information consisting of both the data and knowledge. The data gathered by sensors are processed by knowledge incorporated in the software modules. To explore both the data and knowledge requires to integrate both of them on appropriate levels. As the data usually represent the same part of the world it is much more precise to speak about their fusion than integration. On the other hand, the software modules are usually highly specialized and complementary each to the others. Solving the problems of relevant data fusion and systematic software integration seems to be crucial aspect of the robot design task. Our paper is about the current methods and strategies of data fusion and software integration methods applicable in robotics. Some basic experience gathered during experiments with the GL-robot (the Gerstner Laboratory Robot) experimental platform is presented.