Novel L+ and FO L+ algorithms based on ENC data for automatic route planning of ships

• Published in: Ocean Engineering Vol. 235; p. 109389
• Main Authors: Liang, Cailei, Zhang, Xianku, Watanabe, Yutaka, Zhao, Baigang
• Format: Journal Article
• Language: English; Japanese
• Published: Elsevier Ltd 01.09.2021; Elsevier BV
• Subjects: ENC data; FO L+ algorithm; L+ algorithm; Route planning; FO L+ algorithm; Route planning; L+ algorithm; ENC data
• ISSN: 0029-8018
• DOI: 10.1016/j.oceaneng.2021.109389

This study investigates the problems in the automatic route planning method, which is a critical function for the safe navigation of manned and unmanned ships. We designed a route planning process by extracting the water depth information, such as the position and depth value of water depth points, using the electronic navigation chart (ENC) data. The discrete water depth points were reorganized to form a Delaunay triangulation (DT) data structure, which is the basis of the DT algorithm. The L+ algorithm was proposed and employed on the DT map to provide an appropriate planning route. Furthermore, the FO L+ (final optimization L+) algorithm was used to design an optimized planning route connected by vital waypoints only. Finally, two closed-loop experiments were performed on an electronic chart display and information system platform to demonstrate the effectiveness of the proposed algorithms. Time cost and length of routes were also given to illustrate differences of them. The results of this study indicated that different route planning approaches can be designed based on different safety depths by using the L+ and FO L+ algorithms based on ENC data, rather than grid maps or other pre-processed maps. •Accuracy: Water depth information is extracted from ENC data directly and employed in algorithms.•Innovation: A new strategy for identifying the adjacent nodes is given in the algorithms. Additionally, navigation safety is guaranteed.•Feasibility: L+ and FO L+ algorithms are applied on ECDIS platform, which demonstrates that the algorithms are feasible for practical use.