Research on the Process-based Generation Method of Scenarios Database for Maritime Autonomous Surface Ship Test

Konferenz: ICETIS 2022 - 7th International Conference on Electronic Technology and Information Science
21.01.2022 - 23.01.2022 in Harbin, China

Tagungsband: ICETIS 2022

Seiten: 8Sprache: EnglischTyp: PDF

Autoren:
Wang, Xinyu (Qinhuangdao Branch China Classification Society, Qinhuangdao, Hebei, China)
Wang, Yihang; Lan, Yaoyao (School of Mechanical Engineering, Yanshan University, Qinhuangdao, Hebei, China)
Ma, ilin (Science & Technology Innovation and Test Center, China Classification Society, Beijing, China)
Bian, Hui (Parallel Robot and Mechatronic System Laboratory of Hebei Province Key Laboratory of Advanced Forging & Stamping Technology and Science, Yanshan University, Qinhuangdao, Hebei, China)
Zhao, Xuan (Science & Technology Innovation and Test Center, China Classification Society Beijing, China)

Inhalt:
In recent years, Maritime Autonomous Surface Ship (MASS) has been developed rapidly and has demonstrated many new capabilities, which poses a challenge to the testing methods of autonomous ships. Currently, traditional test methods is no longer sufficient to support comprehensive testing and verification. For virtual testing, there is still a lack of a reliable virtual test scenario generation process. Therefore, this paper proposes a method for generating virtual test scenarios for autonomous ships. It is based on the scenario generation frame of the intelligent car and comprehensively considers the feature of the operating environment of the MASS. This process develops an element database by deconstructing the real scenes. For generating, it reconstructs and assigns the elements according to the test task (functional scenario) and the logical relationship between the elements (logical scenario) to reconstruct the new scenario. The new scenarios obtained in this way are optimized and supplemented to be the test scenarios database. In addition, a prototype following this frame was designed and loaded with the "automatic pathfinding task of the Wuhan section of the Yangtze River under good navigation conditions" for testing. The experiment generated 48 different scenarios, proving that this generator can output a comprehensive test scenario database to meet test requirements.