Advanced EVPH: Advanced Omni-Directional Vector Polar Histogram for Robust Navigation in Crowded Environments

Abstract

This study introduces a novel approach called the Advanced Omni-Directional Vector Polar Histogram (Advanced EVPH), which significantly enhances the navigational efficiency and safety of the conventional EVPH in crowded environments. The conventional EVPH method faces limitations such as a restricted avoidance radius and susceptibility to deadlock scenarios, compromising reliable navigation. To address these challenges, we propose an improved cost function that incorporates midpoint information from laser point groups with a movement direction weight function. This integration extends the obstacle avoidance radius and effectively resolves the deadlock issues. Moreover, unlike conventional methods, which primarily focus on static and concave obstacles, our approach generalizes the symbol function to effectively include dynamic obstacles and their trajectories. Integrating a Kalman filter for dynamic obstacle tracking enables the accurate estimation of obstacle positions and velocities, resulting in proactive and reliable obstacle avoidance. Two simulations demonstrated that the Advanced EVPH improves navigational safety and reduces traveling time. A real-world experiment was conducted to validate the proposed method's reliable and robust performance in a crowded dynamic environment.