Artificial forces for virtual autonomous ships with encountering situations in restricted waters

ABSTRACT

Even if the same two ships operate in the same encountering situation, the actual strategies and timing of operations may be different. Therefore, the fixed collision avoidance trajectory and ship movements are no longer suitable for simulating the real ship behaviour. This paper tries to develop an artificial intelligent system that fully sensitive to the local circumstances and command a ship in virtual environment. Based on AIS (Automatic Identification System) data, this study has developed artificial forces which help decision makings on-board independently under different situations and catching the stochastic nature of ship behaviours during collision avoidances manoeuvring. Actual ship tracks are helpful for ascertaining the parameters that contribute to artificial forces in collision avoidances, while the correlation coefficient analysis is helpful to distinguish the parameters. This study will help to develop a navigation traffic simulation to reflect the realistic ship behaviour and provide reliable information for port and waterway planning, risk analysis, and mitigation measures. The method can be used in developing algorithms for autonomous ships. The method introduced in this study lays a foundation for developing artificial forces at intersections or bends, although the model is only applicable in straight channels at the moment.     

A new logit‐artificial neural network ensemble for mode choice modeling: a case study for border transport

Logit model is one of the statistical techniques commonly used for mode choice modeling, while artificial neural network (ANN) is a very popular type of artificial intelligence technique used for mode choice modeling. Ensemble learning has evolved to be very effective approach to enhance the performance for many applications through integration of different models. In spite of this advantage, the use of ANN‐based ensembles in mode choice modeling is under explored. The focus of this study is to investigate the use of aforementioned techniques for different number of transportation modes and predictor variables. This study proposes a logit‐ANN ensemble for mode choice modeling and investigates its efficiency in different situations. Travel between Khobar‐Dammam metropolitan area of Saudi Arabia and Kingdom of Bahrain is selected for mode choice modeling. The travel on this route can be performed mainly by air travel or private vehicle through King Fahd causeway. The results show that the proposed ensemble gives consistently better accuracies than single models for multinomial choice problems irrespective of number of input variables. Copyright © 2015 John Wiley & Sons, Ltd.     

汽车驾驶人的可靠性分析

交通运输系统是具有关键性人为差错的冗余系统，通过引入人为差错概念，对时间连续工作驾驶人的可靠性进行分析．用经典的可靠度函数方法或马尔可夫方法建立了同样的驾驶人可靠性数学模型，同时提出了驾驶应力（忧虑）的新概念，指出它是影响驾驶工作工效和可靠性的重要因素。     

A Novel Evolutionary Feedforward Neural Network with Artificial Immunology

A hybrid algorithm to design the multi-layer feedforward neural network was proposed. Evolutionary programming is used to design the network that makes the training process tending to global optima. Artificial im-munology combined with simulated annealing algorithm is used to specify the initial weight vectors, therefore im-proves the probabiligy of training algorithm to converge to global optima. The applications of the neural network in the modulation-style recognition of analog modulated rader signals demonstrate the good performance of the network.     

人工智能在舰船避碰领域的应用

讨论人工智能技术的分类,分析模糊逻辑、神经网络(模糊神经网络)、专家系统、主体(A-gent)和基于案例推理(CBR)技术等各种人工智能技术在避碰领域的研究现状,展望了人工智能技术在避碰领域的发展前景.     

变权值加快收敛的路径寻优实时算法

为获得满意解为目标的最优路径选择问题，给出了一种加权的LRTA^*(Learning Real-TimeA^*)算法，通过改变估价函数值更新规则与解时间和解质量的相对折中，加快算法收敛速度。实例应用表明，该方法比LRTA^*算法更快地收敛于满意解，是一种求解大城市稠密路网两点间最优路径的有效方法。     

基于知识的智能CAD系统设计

通过对工程智能ＣＡＤ系统及其设计模型的研究,提出了工程智能ＣＡＤ系统设计理论;规划控制下的二阶段设计理论,论述了基于这一理论的工程智能ＣＡＤ系统设计方法、设计对象的模型构造方法和智能ＣＡＤ系统中各类工程设计知识的应用。并已将这一方法成功地应用于工程智能ＣＡＤ系统的开发实践。     

自组织人工神经网络在民航旅客盈利能力细分中的应用

目前采用的基于里程的旅客划分方法不能清晰区分旅客价值，应用Kohonen T自组织特征映射的人工神经网络模型将反映旅客盈利能力的多维行为特征属性数据以有序的方式映射到对旅客盈利性判别等级的低维空间，形成对旅客正确识别的一种拓扑意义的有序图。试验结果表明，此模型对客户细分识别成功率较高，准确率可达90．8％。     

改进动态克隆选择算法在邮件过滤中的应用研究

提出了可控变异和随机变异方法,改进了动态克隆选择算法,建立并实现一种改进的基于人工免疫系统的邮件过滤算法.并采用SpamAssassin垃圾邮件样本对算法性能进行测试.实验结果表明,引入改进的动态克隆选择算法后,该算法对具有变异特性的垃圾邮件能保持较高的正确识别率,具有较好的自适应性和多样性.     

习题库与指导方法模型的设计与实现

本文研究了智能计算机辅助教学系统中习题库的实现方法，智能指导方式的设计思想及状态信息数据加密解密的保存。