Multimodal gaze shift model using vision and auditory information for humanoid robot

Humanoid robots, which are made to resemble people, are essential for helping people with a variety of jobs. By creating a system that enables a humanoid robot to adjust its gaze to significant locations in its surroundings utilizing both aural and visual inputs, our goal in this work is to improve human-robot interaction. To develop a system that effectively blends acoustic and visual information, we take inspiration from disciplines such as human-computer interaction, machine learning, and neuroscience. Our method uses a neural network type of computer program to create a "map" that highlights important places depending on what the robot sees and hears. The robot's eyes follow this map to determine where to gaze. To assess how well our system performs. Through this research, our goal is to improve human-robot interaction and collaboration, resulting in more seamless and productive exchanges.