Semantic Perception using Spatial Potential Fields

Abstract :

Dealing with knowledge from the human natural environment is one of the most important capabilities, which mobile robots should be equipped with to act autonomously. Moreover, the robot cannot perform a given task correctly without knowing the meaning of the objects involved in this task. Furthermore, semantic perception is very challenging, because the robot must deal with uncertainties and ambiguities that may occur partially caused by noisy sensors. In this paper the concept of novel approach for semantic object recognition for indoor scenes is presented. In our method spatial potential fields are defined and used to recognise a specific object and find its most probable position in a given environment. In addition, we propose a method, in which probability can be applied to deal with missing information and insufficient knowledge.