Robot Visual Tactile Information Fusion Drives Visual Attention

Abstract

This paper describes a theoretical model of visual tactile information fusion mode to drive visual attention. According to the visual and tactile information obtained by the robot, the corresponding position of the most important object and the stimulation signal point for the current task is determined, and the visual attention is focused on the target object of interest. According to the error information fed back by the sensor, the relevant parameters are continuously corrected, and the robustness of the system is improved. The theory was tested on a robot with two degrees of freedom on the head and two degrees of freedom on the arm. The test results show that the model can adapt to changes in itself and the environment and has good robustness.