Dynamic Proxemic Distances in Different Soical Environments

Traditional robots are widely used in industrial which is isolated with humans to keep safety. However, with the advance in mobile robotics, they are going to live or work with human beings and increasingly appear in social environments. When navigating in a shared environment, a mobile robot should not only pay attention to obstacle avoidance and keep safety, it should also consider whether it interferes with other people's personal space and keep the human beings comfortable. The social aware robot has been of more interest to the robotics community. In this paper, we present a method to adjust the proxemic distance dynamically to meet the harmonious coexistence between humans and robots with the change of the social scenes. We combine the semantic information with human movements to generating an appropriate proxemic distance for pedestrians. In contrast to most existing methods, not only security regions but also human psychological characteristics are considering in this method.