E-Planner: An Efficient Path Planner on a Visibility Graph in Unknown Environments

Autonomous navigation in unknown environments holds significance for mobile robots. This article presents a novel efficient path planner, named E-Planner, which utilizes the visibility graph by the obstacle contour optimization and prioritized exploration mechanism. The E-Planner is designed for car-like mobile robots operating in unknown environments. In order to enhance the efficiency of constructing the visibility graph, an obstacle contour optimization strategy is devised to eliminate unnecessary vertices and edges. Furthermore, a breadth-first search algorithm is employed along with prioritized exploration and a timer mechanism to reduce the wastage of exploration space during the path searching process. Additionally, a pure tracking method based on dynamic sight distance is proposed to fulfill the kinematic constraints for Ackermann steering. To validate the effectiveness of these innovative approaches, simulations, and experiments were conducted. The results demonstrate that the proposed obstacle contour optimization strategy significantly reduces the time required for path search. Our method is mostly 25.3% faster than the FAR Planner, and the path is mostly 39.9% shorter, indicating substantial real-time potential. Further research could explore the applicability of these methods to other robotic systems.