A fuzzy decision making algorithm for safe driving in urban environment

Safe and autonomous driving is one of the most important challenges in mobile robotics and has been received considerable attention over the past years in indoor and outdoor navigations. Most methods developed so far immediately activate an obstacle avoidance algorithm when a robot meets obstacles without predicting the motion of the obstacle. These methods would be inefficient for the navigation in urban environments with traffic lane because the traffic lane becomes a constraint in the robot motion. For the safe driving in urban environments, it is efficient to consider this constraint before performing an obstacle avoidance algorithm in the planning phase when the robot meets an obstacle. Therefore, a decision making algorithm for safe driving in case of navigating on the road is needed. In terms of its simplicity and its short response time, a fuzzy algorithm is especially suitable for real-time applications. In this paper, we propose a fuzzy-based decision making algorithm for the outdoor navigation of mobile robots. The algorithm is tested in crossroad environment. To satisfy the robot's safety requirements and to spend less time on the intersection, we designed our navigation algorithm consists of two primary parts: perception (understanding environment) and decision making part. This paper focuses on the decision making part. Simulation results show the algorithm's effectiveness.