Ship behavior intention recognition and reasoning method and device fusing COLREGs scene knowledge

Abstract

The invention discloses a ship behavior intention recognition and reasoning method and device fused with COLREGs scene knowledge. The method comprises the steps of obtaining ship navigation data; classifying the ship behavior intentions based on COLREGs to obtain a ship behavior intention set; establishing a ship behavior first-order logic predicate knowledge base fused with COLREGs according to the ship behavior intention; establishing a ship behavior rule knowledge base based on the established ship behavior first-order logic predicate knowledge base; establishing a Markov logic network based on the established ship behavior rule knowledge base; and identifying and reasoning behaviors of the ship according to the input navigation data and the established Markov logic network. According to the ship behavior intention semantic recognition and reasoning method, the problem that context background knowledge is lacked in existing ship behavior intention recognition and reasoning can be solved, physical significance can be given to ship navigation data from the level of human brain cognition by fusing COLREGs scene knowledge, semantic recognition and reasoning of the ship behavior intention are achieved, and the ship behavior intention is accurately recognized.

Description

technical field [0001] The invention belongs to the technical field of ship behavior intention recognition, and in particular relates to a ship behavior intention recognition and reasoning method and device integrating COLREGs scene knowledge. Background technique [0002] With the continuous development of global trade, the flow of water traffic is becoming hectic and complicated. At the same time, the shipbuilding industry is gradually becoming larger and faster. These factors bring a lot to the supervisors and drivers of water traffic. One of the issues worth noting is to obtain the intention of the ship's navigation behavior, on the one hand, it is helpful for the water traffic supervisor to understand the water traffic situation in time to make corresponding management measures, and on the other hand, it is helpful for the ship's driving guidance. For personnel, they can grasp the sailing intentions of surrounding ships and provide meaningful reference for the safe navi...

AI Technical Summary

Problems solved by technology

However, the above three methods have certain defects. First, most statistical methods use Gaussian mixture models, kernel density estimation and other statistical methods to study ship trajectory data. Such methods need to collect massive ship navigation data, and There are many parameters that need to be estimated, and the amount of calculation is large; second, methods based on machine learning often use clustering, classification, neural network and other methods to study ship navigation data, but such research is more targeted and often applicable In a specific water area, and based on statistical methods and machine learning methods, the focus is on the data layer, often ignoring the influence of the ship's navigation environment on ship behavior, which is the semantic layer of ship behavior; third, knowledge-based Modeling methods mostly use OWL, prolog and other languages ​​to formalize ship behavior, which can reflect the influence of navigation environment on ship behavior from the semantic level. This kind of method pre-sets ship behavior knowledge base, but these prior knowledge bases are difficult to adapt to In the uncertain water traffic environment, it is difficult to accurately identify the behavior intention of the ship, which leads to a high rate of false alarms

Finally, the above three types of methods do not take into account the influence of COLREGs on ship behavior intentions. The semantic layer of ship behavior includes not only the geographical environment of navigation, but also the constraints of water traffic rules on its behavior. Therefore, it is necessary to incorporate the relevant knowledge of COLREGs into ship behavior. Behavior Intent Recognition

Images