Search

We present an improved kernel recursive least squares (KRLS) algorithm for the online prediction of nonstationary time series. In order to adaptively sparsify a selected kernel dictionary for the KRLS algorithm, the approximate linear dependency (ALD) criterion based KRLS algorithm is combined with the quantized kernel recursive least squares algorithm to provide an initial framework. In order to sufficiently track the strongly changeable dynamic characteristics due to nonstationarity, a forgetting factor is further inserted into the proposed combined algorithm. It is shown that our proposed algorithm, referred to as the FFIKRLS algorithm, provides a clearly compatible algorithm structure and can be improved by the existing modeling techniques from both mapping and weights updating perspectives. Numerical simulations using benchmark Lorenz time series in comparison with existing methods have demonstrated that the proposed algorithm has superior performance in terms of both predictive accuracy and kernel dictionary size.

Advancements in complexity, complex systems, and the intelligence sciences, particularly smart city technologies, have shown great potential in aiding to ease traffic congestion. The overall approach and the main ideas in building smart transportation for smart cities, particularly ACP (artificial system, computational experiment, and parallel execution)-based parallel transportation management and control systems (PTMS), are presented. PTMS can be expanded to the new generation of intelligent transportation systems. The main components of the proposed architecture include social signal and social traffic, ITS clouds and services, agent-based traffic control, and transportation knowledge automation. Some technical details of these components are discussed. Finally, one case study is introduced, and the effectiveness is analyzed.

Big data for social transportation brings us unprecedented opportunities for resolving transportation problems for which traditional approaches are not competent and for building the next-generation intelligent transportation systems. Although social data have been applied for transportation analysis, there are still many challenges. First, social data evolve with time and contain abundant information, posing a crucial need for data collection and cleaning. Meanwhile, each type of data has specific advantages and limitations for social transportation, and one data type alone is not capable of describing the overall state of a transportation system. Systematic data fusing approaches or frameworks for combining social signal data with different features, structures, resolutions, and precision are needed. Second, data processing and mining techniques, such as natural language processing and analysis of streaming data, require further revolutions in effective utilization of real-time traffic information. Third, social data are connected to cyber and physical spaces. To address practical problems in social transportation, a suite of schemes are demanded for realizing big data in social transportation systems, such as crowdsourcing, visual analysis, and task-based services. In this paper, we overview data sources, analytical approaches, and application systems for social transportation, and we also suggest a few future research directions for this new social transportation field.

This article presents a new idea by applying different sound package parameters of to improve the acoustic performance of the commercial vehicle cab. A statistical energy analysis model of the commercial vehicle cab was created, and the SPL of the interior noise for the commercial vehicle cab was predicted and validated at speed of 80 km/h. The contributions of some major subsystems to interior noise were obtained at 80 km/h. Two major characteristics of interior SPL and total weight of sound packages were determined as the optimization targets. The thicknesses of the sound packages for floor, roof, front wall and rear wall were selected to be the optimal variables. The orthogonal array table with four factors and three levels are selected for the experimental design. The optimal combination of sound package parameters is determined by using grey relational analysis with Taguchi method. A grey relational grade is acquired from the analysis to optimize multiple performance characteristics. The experimental results show that the presented methods can effectively deal with the information redundancy problem with multiple sound package parameters for commercial vehicle cab.

A cyber-physical system (CPS) is composed of a physical system and its corresponding cyber systems that are tightly fused at all scales and levels. CPS is helpful to improve the controllability, efficiency and reliability of a physical system, such as vehicle collision avoidance and zero-net energy buildings systems. It has become a hot R&D and practical area from US to EU and other countries. In fact, most of physical systems and their cyber systems are designed, built and used by human beings in the social and natural environments. So, social systems must be of the same importance as their CPSs. The indivisible cyber, physical and social parts constitute the cyber-physical-social system (CPSS), a typical complex system and it’s a challengeable problem to control and manage it under traditional theories and methods. An artificial systems, computational experiments and parallel execution (ACP) methodology is introduced based on which data-driven models are applied to social system. Artificial systems, i.e., cyber systems, are applied for the equivalent description of physical-social system (PSS). Computational experiments are applied for control plan validation. And parallel execution finally realizes the stepwise control and management of CPSS. Finally, a CPSS-based intelligent transportation system (ITS) is discussed as a case study, and its architecture, three parts, and application are described in detail.

A deep understanding of people's activity-travel behaviors is critical and essential for effective travel demand forecasting and management. Although it is acknowledged that social interactions play an important role in people's decision-making behaviors, our understanding of how they shape and impact activity-travel behaviors of people is still limited. Therefore, for the first time, this paper introduces social learning into artificial transportation systems (ATSs) to model their influence on activity-travel behaviors. Based on a specified ATS, three types of universal social interactions (i.e., imitation, conformity, and experience sharing on social networks) are modeled and studied. The results indicate that our models can make artificial agents learn to decide the best behavior, form habitual choices, and emerge fashion gradually.

Social activity-travel has gained more and more attention as it is a growing percentage of the whole travel. To study its generation mechanism and behavioral characteristics, social network data are usually essential. However, due to individual privacy, it is rather difficult for traditional methods such as questionnaires to collect abundant reliable data. Therefore, we propose a novel method to grow realistic social networks based on artificial transportation systems (ATS). By incorporating the activity-travel simulation provided by ATS and a new agent-based model for social interaction, the method takes into account human mobility to generate spatially embedded social networks. Human mobility shapes and impacts social networks dynamically but is usually ignored by related studies. A case study based on computational experiments is carried out to verify the method. The results indicate that the method can generate social networks with similar topological and spatial characteristics to real social networks.

This paper proposes a method based on the Artificial societies, Computational experiments, and Parallel execution (ACP) approach to build parallel public transportation systems (PPTSs). The framework and components of a PPTS are analyzed, and some details for building the PPTS are discussed. One prototype based on intelligent traffic clouds is established. One specific PPTS is developed for the Guangzhou 2010 Asian Games in the case study, and its effectiveness is verified through the evaluation of two evacuation plans for the Asian Games.

The Artificial societies, Computational experiments, and Parallel execution (ACP) approach provides us an opportunity to look into new methods that address transportation problems from new perspectives. In this paper, we present our work and results of applying the ACP approach on modeling and analyzing transportation systems, particularly carrying out computational experiments based on artificial transportation systems (ATSs). Two aspects in the modeling process are analyzed. The first is growing an ATS from the bottom up using agent-based technologies. The second is modeling environmental impacts under the principle of “simple is consistent.” Finally, three computational experiments are carried out on one specific ATS, i.e., Jinan-ATS, and numerical results are presented to illustrate the applications of our method.

Artificial Transportation Systems (ATS) can support a variety of computational experiments for different purposes, and have become important tools for transportation research. However, when the road network modeled in ATS is large or there are a lot of vehicles, ATS will run very slow or even cannot be started due to the memory limit of a standalone computer. With the acceleration of urbanization process and rapid increase of car ownership, it is necessary to find a high-performance computing method for running large-scale ATS. Therefore, the paper presents an Erlang-based approach to realize concurrent and distributed computing of ATS. To verify the feasibility, a prototype is built, and a performance test is conducted. The results show the method can achieve the efficiency utilization of computing resources.

On the basis of the introduction of related concepts and theories, analyze the construction process of parallel transportation system and its application in traffic signal control system. The parallel transportation system of Tailiu road is established and applied in the estimation of effectiveness, the selection of control plan, and the development trend forecast in the project of rebuilding traffic signal control system on Tailiu road. It is shown that our methods are effective in practice.

There has been an ever-increasing interest in research on traffic detection and traffic scene understanding based on computer vision, driven by the increasing processor speed and the increasing deployment of cameras. This field is of increasing relevance for the intelligent transport system (ITS). Therefore, there seems to be a real need for an analytical review of recent developments in this domain. We present here is a comprehensive review of video processing techniques state of the art for vehicle detection and tracking and an outlook to future research directions.

Evacuees' behaviors have a significant impact on evacuation efficiency. Evacuation clearance time is one of the key indicators in the evacuation planning and management. Evacuees' departure time choice, destination choice and route choice behavior based on real-time traffic information are three crucial factors used to estimate evacuation clearance time. In this paper, we use the orthogonal experimental design method to determine values of these three factors simultaneously, which can reduce evacuation clearance time significantly. Once we obtain these values, we can guide evacuees' departure time choice, destination choice and route choice behavior to approximate these values and evacuation efficiency will improve greatly. The dynamic traffic evacuation process is modeled using a rule-based multi-agent microscopic traffic simulation system called TransWorld. The simulation experiment results illustrate the effectiveness and reliability of the proposed method. The proposed methodology, computational results and discussions can be used for future emergency evacuation planning and management.

ACP (Artificial societies, Computational experiments, and Parallel execution) approach is adopted by our method to build parallel public transport system (PPTS). The framework of PPTS is proposed and some components are analyzed. Two key steps in building up PPTS are discussed. The first is growing up artificial public transport system from bottom up using agent-based technology. The second is implementing schedule plans of public vehicles using cloud computing. One specific PPTS is established for Guangzhou 2010 Asian Games. Its effectiveness is verified and illustrated by comparing the traffic parameters between before and after the employment.

Information of Activities location is critical for traffic simulation based on activity. To replace time-consuming manual acquisition, this paper proposes an information extraction method based on ontology and frequent subtree to acquire place-related information from the Web automatically. A destination selection model based on Nested Logit is established to use this information in traffic simulation. To verify the method and model, information of places within Zhongguancun area is acquired from the Web, and then experiments with uniform design are carried out on artificial transportation systems and regression analysis is done.

Artificial Transportation Systems (ATS) provide a comprehensive perspective to study actual transportation systems, which are a kind of open and complex giant system referring to diverse engineering and social disciplines. To make use of achievements from related research fields, ontology for ATS is constructed in this paper. Besides conceptual model involved in modeling and applications of ATS, the compatibility with other ontology is considered. Further, an ontological knowledge base with SOA (Service-Oriented Architecture) is designed to manage information and knowledge in ATS. The architecture and interfaces of the knowledge base are presented. Benefit from this, construction and applications of ATS can be conducted on a uniform information and knowledge platform. A case study about traffic jam pattern is demonstrated.

In this chapter, a concept framework for urban traffic management system based on the ontology and multiagent system technology is proposed. The framework has advantages such as robustness, flexibility, and scalability. Particularly, the traffic management system ontology is analyzed in detail. It is classified into seven classes, and each of them is further investigated. Then, the three-leveled multiagent architecture for the implementation of the management system is briefly explained.