Perception-Aware A* Path Planning for Range-based Navigation

Abstract

The mobile robots path planning mostly considers metrics of minimizing path length, minimizing time, or optimizing smoothness. However, the perceptual information gain is also important for the scenes requiring high state estimation accuracy. This paper proposes a perception-aware A* path planning algorithm aiming at improving the information captured by Lidar. Firstly, the Lidar-scanning model for specific nodes in the obstacle map is constructed. Then, by quantifying the information gain of the node, a new cost function considering perception is formulated, and the perception-aware A* algorithm is proposed for mobile robot path planning using range-based navigation. Finally, the effectiveness and performances of the proposed algorithm are verified through simulation experiments.