Anticipatory Robot Navigation: Incorporating Estimated Obstacle Behaviors with the Social Force Model

In environments populated by various dynamic obstacles, such as people and robots, it is essential for each robot to operate efficiently while avoiding these obstacles. Therefore, in this paper, we propose a method for navigating a robot to avoid obstacles and ensure smooth and efficient movement in dynamic environments. Moving obstacles may also actively attempt to avoid the robots. Therefore, our navigation method should consider such interactions to achieve an efficient movement. To achieve this, our method integrates principles from the Social Force Model and models the movement of nearby obstacles by estimating the parameter of the Social Force Model. Then, our method controls the robot to ensure a balance between efficiency and safety based on the model. We evaluate our method by simulation and demonstrate that it achieves high efficiency and safety by utilizing a model of nearby obstacles.