Sensor observation area compensating path planning for avoiding collisions with unknown obstacles

Abstract

In this paper, we propose a path-planning algorithm for mobile robots having sensors with a limited field of view. The proposed method predicts the observable region in the planning phase and minimizes incursions into unobservable regions. The path-planning process of the proposed method not only involves the actions required for reaching a destination but also the actions required for obtaining information about obstacles on the path depending on the sensor's viewing angle. Using the proposed method, robots can detect obstacles before collision. In this paper, the effectiveness of the method is confirmed by simulated and real-world experiments.