Multi-density Clustering Based Hierarchical Path Planning
Tianyi Bai, Shihua Yuan, Xueyuan Li, Xufeng Yin, Junjie Zhou · 2019
Path planning is one of the key abilities in artificial intelligence. Many different approaches exist, focusing on finding the shortest path. Obstacle density, which is often ignored by conventional approaches, has a great influence on path quality especially in off-road environment. Because obstacles along a path compromise driving safety and frequent obstacle avoidance increases vehicle control difficulty while decreases traveling speed. In this paper, a simple and efficient approach of hierarchical path planning algorithm based on multi-density clustering is proposed, aiming at finding a clear and short path to achieve an integrated goal of obstacle avoidance and path length shortness. A simulation evaluation shows that our proposed approach can find short path bypassing regions with dense obstacles efficiently.