A Computing Model of Selective Attention for Service Robot Based on Spatial Data Fusion

Abstract

Robots and humans are facing the same problem: they all need to face a lot of perceptual information and choose valuable information. Before the robots provide services, they need to complete a robust real-time selective attention process in the domestic environment. Visual attention mechanism is an important part of human perception, which enables humans to select the visual focus on the most potential interesting information. It also could dominate the allocation of computing resource. It also could focus human’s attention on valuable objects in the home environment. Therefore we are trying to transfer visual attention selection mechanism to the scene analysis of service robots. This will greatly improve the robot’s efficiency in perception and processing information. We proposed a computing model of selective attention which is biologically inspired by visual attention mechanism, which aims at predicting focus of attention (FOA) in a domestic environment. Both static features and dynamic features are composed in attention selection computing process. Information from sensor networks is transformed and incorporated into the model. FOA is selected based on a winner-take-all (WTA) network and rotated by inhibition of return (IOR) principle. The experimental results showed that this approach is robust to the partial occlusions, scale-change illumination, and variations. The result demonstrates the effectiveness of this approach with available literature on biological evidence. Some specific domestic service tasks are also tailored to this model.