Cognitive computing models offer a formal and interpretable way to characterize human's deliberation and decision-making, yet their development remains labor-intensive. In this paper, we propose NL2CA, a novel method for auto-formalizing cognitive decision-making rules from natural language descriptions of human experience. Different from most related work that exploits either pure manual or human-guided interactive modeling, our method is fully automated without any human intervention. The approach first translates text into Linear Temporal Logic (LTL) using a fine-tuned large language model (LLM), then refines the logic via an unsupervised Critic Tree, and finally transforms the output into executable production rules compatible with symbolic cognitive frameworks. Based on the resulted rules, a cognitive agent is further constructed and optimized through cognitive reinforcement learning according to the real-world behavioral data. Our method is validated in two domains: (1) NL-to-LTL translation, where our CriticNL2LTL module achieves consistent performance across both expert and large-scale benchmarks without human-in-the-loop feedbacks, and (2) cognitive driving simulation, where agents automatically constructed from human interviews have successfully learned the diverse decision patterns of about 70 trials in different critical scenarios. Experimental results demonstrate that NL2CA enables scalable, interpretable, and human-aligned cognitive modeling from unstructured textual data, offering a novel paradigm to automatically design symbolic cognitive agents.

Ergonomics or human factors engineering (HFE) mainly exploits human experiments to discover one's cognitive and behavioral mechanisms. Such a paradigm, however, suffers from the scale of subject group and the extent to which they can stand for the whole studied population. Additionally, for real-time human-machine tasks, the experiment-modeling-validation-application path may not be applicable since the experiment cannot be flexibly conducted to update cognitive models, leading to a failure of the online system control and management. To solve the dilemma, this article proposes the generative artificial intelligence (GAI)-driven ergonomics to augment the HFE research. By introducing GAI techniques, virtual-real hybrid experiments are combined and supplement more heterogeneous samples, enhancing the input diversity for cognitive modeling and behavioral learning. The case studies of human-machine cooperative driving and aerospace robotic arm operation indicate that the innovative paradigm can effectively and efficiently augment the human experiment data. It can elevate the generality and robustness of human models.

Typical human resource management relies on equilibrium analysis or manual design to develop appropriate strategies. Such a paradigm, however, may not be applicable for modern enterprises, as systems are highly time variant with diverse participants. To tackle such individual cognitive difference and systemic complexity, this article proposes a distributed model repository—the digital personnel with heterogeneity. The digital personnel are composed of different digital employers and employees and can conveniently conduct computational experiments to dynamically explore possible management policies. By cognitive reinforcement learning and constant interaction with the real human users, it provides a knowledge automation approach for the agile reengineering and optimization of workflows as well as the elevation of user's satisfaction. Experiment based on a multinational company manifests that the methodology is feasible and effective. Consequently, the heterogeneous digital personnel play a source of intelligent behavior generation with interactive computational experiments and can be an efficient testbed for management strategies.

Traditional cognitive travel modeling typically employs a unified cognitive model to simulate representative travel behaviors, which may usually result in a weak characterization of user heterogeneity in paths, modes, and other factors. Large language model (LLM), by contrast, has significantly enhanced the anthropomorphic and personalized features of intelligent systems. To integrate their advantages, this article proposes LLM-driven cognitive modeling to generate more diverse and personalized travel demands. The new method sufficiently exploits LLM such as the llama as a basis and provides personalized travel plans so that more heterogenous travel demands could be generated. Additionally, introducing LLM into cognitive modeling can significantly reduce the time of model development, thus accelerating the research or engineering deployment. By calibrating and testing with one month’s data from public transportation (buses and subways) in Beijing, our method, compared to traditional cognitive models, not only achieves better accuracy in reproducing typical travel patterns, but also generates more diverse ones, providing a more comprehensive input for computational experiments on traffic management and control strategies.

With the rapid development of autonomous driving technology, the safety of driving systems has increasingly become the focus of attention. However, although many existing autonomous driving decision-making algorithms, such as deep reinforcement learning, demonstrate excellent performance, their decision-making processes lack interpretability and are opaque to users. To address this problem, this paper constructs an interpretable driving model from the perspective of human cognition, which can not only imitate human driving behavior through cognitive reinforcement learning methods, but also show better performance in driving experiments. In addition, the paper also proposes an analysis method for abnormal driving behavior, which provides a new idea for discovering potential unsafe behaviors during driving and exploring the possible impact of this behavior pattern on driving tasks.

Accurately forecasting the future paths of numerous agents is vital for the efficacy of autonomous systems. In crowded scenarios such as sidewalks, subways and airports, pedestrians instinctively modify their motion pattern in response to the environmental context and social consensus like preserving personal space and circumventing physical contact. Thus, the task to predict future pedestrian trajectory presents considerable challenges owing to the complex interaction among agents and the inherent uncertainty in predicting each agent's subsequent actions. Inspired by the recent success of Graph Neural Networks (GNN), a model named DATraj is introduced for predicting pedestrian trajectory. DATraj first uses a temporal encoder composed of attention mechanism to capture the spatial-temporal dynamics of pedestrians. The encoder can learn the motion pattern and subtle movement of pedestrian in the crowded scenario. Graph Attention Networks (GAT) is used in many models to catch social interaction between individuals. However common GATs compute a static attention: the ranking of the attention scores is unconditioned on the query node. DATraj implement the global interaction parts using the improved dynamic attention which every query uniquely prioritizes the attention coefficients correlated with the keys, this provides a much better robustness to noise. Experiments show that our trajectory prediction model achieves better performance on several public datasets.

The coronavirus disease 2019 (COVID-19) pandemic has changed the world deeply. Urban trajectory big data collected by wireless sensing devices provide great assistance for COVID-19 prevention. However, except for contact tracing, trajectory data are rarely employed in other preventative scenarios against the pandemic. In this article, we try to extend the application of trajectories auto-collected by wireless sensing devices and simulate the epidemic spread in a trajectory data-driven manner. After that, the effects of three nonpharmacological measures are quantified. In contrast to existing studies, additional requirements such as the complex topological networks are needless in our simulation, where the interactions between agents are derived by the intersections of their trajectories. Concretely, the dynamic of virus propagation among individuals is first modeled, and then an agent-based microsimulation environment is built as an artificial system to conduct the epidemic spread simulation. Finally, the trajectories are loaded into the agents as the reliance for their interactions, and the macroscopic changes under different interventions are revealed in a bottom–up way. As a case study, we conduct the simulation based on the trajectories in a real region, in which we find the following. 1) Among the three examined nonpharmacological interventions, community containment is more effective than keeping social distance, which can lower the deaths to nearly 1/9 compared to no action, while travel restrictions play limited roles. 2) There is a strong positive correlation between population densities and mortality. 3) The timing of community containment triggered by confirmed diagnoses is proportional to the number of deaths, thus early containment will significantly decrease mortality.

Digital human in cyberspace can help provide humanized services in specific applications, such as question & answer systems, recommender systems, chatter robots, and intelligent assistants. While most researches focus on behavior analytics, few of them integrate the personality that is also a closely related factor. As a classic indicator for personality representation, Myers–Briggs type indicator (MBTI) categorizes an individual into mutually exclusive types from four dichotomous axes (extraversion versus introversion, sensing versus intuition, thinking versus feeling, judging versus perceiving). Traditional recognition method using MBTI simply measures the user’s preference frequency in each axis through questionnaires, treating the dominant value as the identified result. Such a paradigm, however, represents all the people with only 16 types and cannot distinguish heterogeneous users clearly. This article proposes a novel personality recognition method using fuzzy logic. Different from previous classifications, our new method categorizes the individual in a continuous space and represents one’s personality in a more fine-grained level. We have designed comparative psychological tests for 77 people. The validation experiments on such tests indicate that the fuzzy-logic-based method is not only consistent with the classic MBTI tests (in the sense of defuzzification) but also provides the uncertainty for each personality type. Therefore, it can be viewed as a generalization of the classic MBTI tests and promotes the representation of individual’s heterogeneity for fine-grained analytics of digital human.

The evolution of modern mobile terminals, social networks, and other intelligent services makes everyone become a ubiquitous information perceiver, producer, and propagator. Also known as “social sensor” and “social IoT,” these individuals and communities generate a huge volume of social signals, which has shown prominent value for mining. These unstructured social signals provide a new perspective in the research of complex systems, which makes the traditional cyber–physical system (CPS)-oriented information computing sublimate to the cyber–physical–social system (CPSS)-oriented knowledge computing. However, there still exist great uncertainties, ambiguities, and complexities in modeling behaviors of social individuals or groups. Especially when we apply big-data-driven learning-based models in specific fields and scenarios, the lack of domain expert knowledge and characteristics of system uncertainty severely limits the performance and accuracy of these models. The introduction of fuzzy system modeling integrates data and knowledge in the social computing area, which has shown its unique advantages in solving the above issues and has drawn more attention to this topic. In this article, we conduct a review of recent advances in social computing with fuzzy technologies in CPSS. First, we briefly review the development of social computing, and analyze the characteristics and advantages of social computing through fuzzy methods. Second, we refine core fuzzy system methods for social computing and elaborate on existing fuzzy-technology-empowered social computing methodologies. As in a range of social spaces, we also review and analyze related advances in human-in-the-loop systems. We also reveal the trend of decentralized, autonomous, and organized computing in cyber–physical–social space with fuzzy-based methods and proposed a framework to categorize related studies in CPSS. Finally, we conclude the research trends and hotspots based on current studies, and discuss the challenges for future research directions.

Affected by people’s dynamic social activities, the imbalance between vehicle supply and demand in the Mobility-On-Demand(MOD) system is a common phenomenon. To improve traffic efficiency, an Auxiliary Network Enhanced Hierarchical Graph Reinforcement Learning (AHGRL) method is proposed for vehicle repositioning. Firstly, a hierarchical graph reinforcement learning (HGRL) framework is designed. The complex vehicle repositioning problem in real road networks is divided into many sub-tasks and multiple reinforcement learning algorithms are designed to solve decision problems of different levels. Traffic congestion is also considered and road nodes are clustered dynamically. And then an auxiliary graph reinforcement learning (AGRL) algorithm is designed for the actuator. It contains the prediction branch and the repositioning branch. States and rewards of agents could be designed accurately with the support of the prediction branch. The two branches cooperate in an auxiliary way to achieve excellent forecasting and repositioning effects. Finally, to enable efficient multi-vehicle coordination, a discrete Soft Actor-Critic algorithm is adopted in the repositioning branch, which learns multiple optimal actions for vehicles in the same area. Comparative experiments with real data demonstrate the effectiveness of our method. And ablation experiments verify the effectiveness and universality of the HGRL framework and the AGRL algorithm.

The era has been witnessing our entering into a more professional and cooperative society and also a bloom of various digital humans that help us complete more challenging industrial work than ever. As a major human-machine communication channel, virtual-real interaction between digital and real humans is an efficient way to adjust the mismatch between specialized tasks and participants with unsuitable personalities, elevating the overall performance of such human participated systems. This article proposes a new paradigm for temporary personality shaping by prescriptively interacting the human operator/user with his “digital assistant” in cyber space. As a first proposed personality shaping with AI-aided approach, the new paradigm involves digital human modeling with uncertain personality, computational experiments on personality evolution, and prescriptive interaction with user counterpart. By iteratively and repeatedly executing the three steps, the personality of human participants is gradually tailored and prescribed so that he can well undertake specific tasks. Experiments based on online social media data and human-machine shared driving have indicated that our new personality shaping paradigm is feasible and effective in performing representative tasks.

Foundation models are used to train a broad system of general data to build adaptations to new bottlenecks. Typically, they contain hundreds of billions of hyperparameters that have been trained with hundreds of gigabytes of data. However, this type of black-box vulnerability places foundation models at risk of data poisoning attacks that are designed to pass on misinformation or purposely introduce machine bias. Moreover, ordinary researchers have not been able to completely participate due to the rise in deployment standards. This study introduces the theoretical framework of scenarios engineering (SE) for building accessible and reliable foundation models in metaverse, namely, “SE-enabled foundation models in metaverse.” Particularly, the research framework comprises a six-layer architecture (infrastructure layer, operation layer, knowledge layer, intelligence layer, management layer, and interaction layer), which can provide controllability, trustworthiness, and interactivity for the foundation models in metaverse. This creates closed-loop, virtual–real, and human–machine environments that provides the best indices and goals for the foundation models, which allows us to fully validate and calibrate the corresponding models. Then, examples of use cases from the automotive industry are listed to provide transparency on the possible use and benefits of our approach. Finally, the open research topics of related frameworks are discussed.

As an interdisciplinary research approach, traditional cognitive science adopts mainly the experiment, induction, modeling, and validation paradigm. Such models are sometimes not applicable in cyber-physical-social-systems (CPSSs), where the large number of human users involves severe heterogeneity and dynamics. To reduce the decision-making conflicts between people and machines in human-centered systems, we propose a new research paradigm called parallel cognition that uses the system of intelligent techniques to investigate cognitive activities and functionals in three stages: descriptive cognition based on artificial cognitive systems (ACSs), predictive cognition with computational deliberation experiments, and prescriptive cognition via parallel behavioral prescription. To make iteration of these stages constantly on-line, a hybrid learning method based on both a psychological model and user behavioral data is further proposed to adaptively learn an individual’s cognitive knowledge. Preliminary experiments on two representative scenarios, urban travel behavioral prescription and cognitive visual reasoning, indicate that our parallel cognition learning is effective and feasible for human behavioral prescription, and can thus facilitate human-machine cooperation in both complex engineering and social systems.

Calibration of agent-based models (ABM) is an essential stage when they are applied to reproduce the actual behaviors of distributed systems. Unlike traditional methods that suffer from the repeated trial and error and slow convergence of iteration, this article proposes a new ABM calibration approach by establishing a link between agent microbehavioral parameters and systemic macro-observations. With the assumption that the agent behavior can be formulated as a high-order Markovian process, the new approach starts with a search for an optimal transfer probability through a macrostate transfer equation. Then, each agent's microparameter values are computed using mean-field approximation, where his complex dependencies with others are approximated by an expected aggregate state. To compress the agent state space, principal component analysis is also introduced to avoid high dimensions of the macrostate transfer equation. The proposed method is validated in two scenarios: 1) population evolution and 2) urban travel demand analysis. Experimental results demonstrate that compared with the machine-learning surrogate and evolutionary optimization, our method can achieve higher accuracies with much lower computational complexities.

Population synthesis is the foundation of the agent-based social simulation. Current approaches mostly consider basic population and households, rather than other social organizations. This article starts with a theoretical analysis of the iterative proportional updating (IPU) algorithm, a representative method in this field, and then gives an extension to consider more social organization types. The IPU method, for the first time, proves to be unable to converge to an optimal population distribution that simultaneously satisfies the constraints from individual and household levels. It is further improved to a bilevel optimization, which can solve such a problem and include more than one type of social organization. Numerical simulations, as well as population synthesis using actual Chinese nationwide census data, support our theoretical conclusions and indicate that our proposed bilevel optimization can both synthesize more social organization types and get more accurate results.

Accurate traffic prediction under various conditions is an important but challenging task. Due to the complicated non-stationary temporal dynamics in traffic flow time series and spatial dependencies on roadway networks, there is no particular method that is clearly superior to all others. Here, we focus on investigating ensemble learning that benefits from multiple base models, and propose a traffic-condition-aware ensemble approach that acts as a decision maker by stacking multiple predictions based on dynamic traffic conditions. To sense traffic conditions, we apply the Convolutional Neural Network (CNN) model to capture the spatiotemporal patterns embedded in traffic flow. Then, the high-level features extracted by CNN are used to generate weights to ensemble multiple predictions of different models. Extensive experiments are performed with a real traffic dataset from the Caltrans Performance Measurement System. We compare the proposed approach with competitive models, including Gradient Boosting Regression Tree (GBRT) model, Weight Regression model, Support Vector Regression (SVR) model, Long Short-term Memory (LSTM) model, Historical Average (HA) model and CNN model. Experimental results demonstrate that our method can effectively improve the performances of traffic flow prediction.

It is indispensable for professional traffic signal engineers to perform manual operations of traffic signal control (TSC) to mitigate traffic congestion, especially with complicated scenarios. However, such a task is time-consuming, and the level of congestion mitigation heavily relies on individual expertise in engineering practice. Therefore, it is cost-effective to learn traffic engineers’ knowledge to enhance the problem-solving skills for a large-scale urban traffic network. In this paper, a data augmented deep behavioral cloning (DADBC) method is proposed to imitate the problem-solving skills of traffic engineers. The method is under a conceptual framework, parallel learning (PL) framework, that incorporates machine learning techniques for solving decision-making problems in complex systems. The DADBC method enhances a hybrid demonstration by exploiting a generative adversarial network (GAN) and then uses the deep behavioral cloning (DBC) model to learn traffic engineers’ control schemes. According to the validation results using the real manipulation data from Hangzhou, China, our method can imitate complex human behaviors in intervening traffic signal control operations to improve traffic efficiency in urban areas.

The imbalance of vehicle supply and demand is a common phenomenon that influences the efficiency of online ride-hailing systems greatly. To address this problem, a three-level hierarchical mixed deep reinforcement learning method (HMDRL) is proposed to reposition idle vehicles. A manager operates at the top level, where action-abstraction is conducted from the time dimension and is adaptive for spatially scalable and time-varying systems. Coordinators locate at the middle level and a parallel coordination mechanism that is independent of the decision order is designed to improve the efficiency of the repositioning. The bottom level is composed of executive workers to reposition vehicles with mixed states and the states contain spatiotemporal information of agents’ neighbor areas. Two reward functions are designed for the manager and the coordinators, respectively, aiming to improve the training effect by avoiding sparse rewards. A simulator based on real orders is designed and HMDRL is compared with six methods. Experimental results demonstrate that HMDRL outperforms all the other methods. In three comparison experiments, the order response rate is increased by 0.62% to 8.29%, 1.5% to 7.78%, 1.18% to 4.75%, respectively.

Taxi demand prediction is valuable for the decision-making of online taxi-hailing platforms. Data-driven deep learning approaches have been widely utilized in this area, and many complex spatiotemporal characteristics of taxi demand have been studied. However, the heterogeneity of demand patterns among different taxi zones has not been taken into account. To this end, this paper explores zone clustering and how to utilize the inter-zone heterogeneity to improve the prediction. First, based on the pairwise clustering theory, a taxi zone clustering algorithm is designed by considering the correlations among different taxi zones. Then, both the cluster-level and the global-level prediction modules are developed to extract intra- and inter-cluster characteristics, respectively. Finally, a Multi-Level Recurrent Neural Networks (MLRNN) model is proposed by combining the two modules. Experiments on two taxi trip records datasets from New York City demonstrate that our model improves the prediction accuracy compared with other state-of-the-art methods.

Strategic traffic management is crucial for combating traffic congestion at the macroscopic level. However, such a field is still relatively unexplored, particularly for microscopic control objects, such as intersections and coordinated intersection groups. This article proposes a human-in-the-loop recommendation system for strategic urban traffic management, which follows an agent-based structure. A regional agent dispatcher is defined to assign agents for operation whenever “operation on-demand” is required. Such a requirement is identified by a daily-dependent operational mode on strategic traffic operations at a control object level. The strategic management scheme for each control object is guided by a strategic agent (customized), which is essentially a deep recommender model with a specific architecture. By featuring the multiagent design, a customized operational scheme can be generated at the intersection level, which instructs the corresponding controller to take specific operations. The utility of the recommendation system is demonstrated via a case study using real-world traffic data. In both offline and online evaluations, the system performs consistently at traffic operational recommendations in different scenarios and has the potential to provide more reasonable traffic operational strategies than a human-operated system.

How to learn effective features from large-scale point clouds for semantic segmentation has attracted increasing attention in recent years. Addressing this problem, we propose a learnable module that learns Spatial Contextual Features from large-scale point clouds, called SCF in this paper. The proposed module mainly consists of three blocks, including the local polar representation block, the dual-distance attentive pooling block, and the global contextual feature block. For each 3D point, the local polar representation block is firstly explored to construct a spatial representation that is invariant to the z-axis rotation, then the dual-distance attentive pooling block is designed to utilize the representations of its neighbors for learning more discriminative local features according to both the geometric and feature distances among them, and finally, the global contextual feature block is designed to learn a global context for each 3D point by utilizing its spatial location and the volume ratio of the neighborhood to the global point cloud. The proposed module could be easily embedded into various network architectures for point cloud segmentation, naturally resulting in a new 3D semantic segmentation network with an encoder-decoder architecture, called SCF-Net in this work. Extensive experimental results on two public datasets demonstrate that the proposed SCF-Net performs better than several state-of-the-art methods in most cases.

Agent-based simulation is a useful approach for the analysis of dynamic population evolution. In this field, the existing models mostly treat the migration behavior as a result of utility maximization, which partially ignores the endogenous mechanisms of human decision making. To simulate such a process, this article proposes a new cognitive architecture called the two-layered integrated decision cycle (TiDEC) which characterizes the individual's decision-making process. Different from the previous ones, the new hybrid architecture incorporates deep neural networks for its perception and implicit knowledge learning. The proposed model is applied in China and U.S. population evolution. To the best of our knowledge, this is the first time that the cognitive computation is used in such a field. Computational experiments using the actual census data indicate that the cognitive model, compared with the traditional utility maximization methods, cannot only reconstruct the historical demographic features but also achieve better prediction of future evolutionary dynamics.

Social relationships have a strong influence on individual travel behavior and, consequently, on travel demand. However, most current literatures on population synthesis, which is the fundamental building block of disaggregated travel demand forecasting and agent-based traffic simulation, only considers the household impact. This paper makes two contributions in this regard. First, a methodological issue is identified: the existence of multiple social relationships (e.g., a dual set of constraints from social institutions or structures) makes it more difficult to generate a consistent synthetic population, meaning that this population satisfies constraints from more than one type of social organizations. A tensor decomposition method is then proposed to generate a consistent population with multi-social relationships. To our knowledge, this is the first time that this type of methodological issue has been addressed. Our sample-based method constitutes an improvement compared to existing approaches in that it can respect constraints from multiple social organizations without reducing accuracy. A numerical test concerning individual, household, and enterprise, using Chinese national population and economic census data, indicates that the new method can lead to stable and relatively small errors in total. The source code is available from https://github.com/PeijunYe/MulSocPopSyn.git.

Welcome to the fourth issue of the IEEE Transactions on Computational Social Systems (TCSS), which includes 16 regular papers and a brief discussion on social computing. We would also like to inform you that IEEE will conduct its regular 5-year review for TCSS at its TAB meeting in November at Boston. Any suggestions for our review report are welcome!

Welcome to the last issue of the IEEE Transactions on Computational Social Systems (TCSS) of this year, with a special focus on “blockchainbased secure and trusted computing for IoT.” Here, we have 18 regular articles and a brief discussion on social intelligence. I would like to take this opportunity to thank and congratulate everyone, especially our editorial board for a great job well done. Looking forward to working with you all in 2020!

Crime scene analysis is a retrospective process from traces to psychology and physiology. It is not only the starting point and foundation of criminal investigation, but also the key part for solving criminal cases. As a typical open complex social system, it has three features, namely, uncertainty, diversity, and complexity, thus making the system modeling a huge challenge. In this paper, we propose the parallel crime scene analysis system based on the artificial societies, computational experiments and parallel execution (ACP) approach, which uses artificial (A) crime scene to describe the basic elements, functions and states of the criminals, computational (C) experiments to compute and predict the different forms of crime scene, and parallel (P) execution to guide or control the evolution of the physical crime process in accordance with the results from the artificial crime scene. First, we propose the concept of parallel crime scene from the perspective of complex system theory and give an overview of its architecture, then we present the construction method of artificial crime scene and the blackboard-based multiagent artificial crime scene analysis system. On this basis, the temporal and spatial interaction models of the criminal subjects are proposed and verified. After that, we introduce the software-defined crime scene analyzing model systematically. The ACP approach sheds light on the intelligent management and control for complex crime scene analysis.

Artificial Society is an analytical foundation of various complex eco- and social systems. Such system is usually implemented via multiagent approach. However, there is no consensus on how to model the agent’s decision-making process, since different application scenarios concentrate on different facets. This, to some extent, hinders model reuse and system integration. This paper proposes a general cognitive architecture that attempts to adapt all the aspects of agent’s decision-making in artificial societies, so that different programs and software can be reorganized and integrated conveniently. To illustrate its implementation, two simulations—emergent evacuation and population evolution—are conducted. These tests clearly show that the proposed architecture is able to support different agent-based models. Problems that might be encountered, as well as possible strategies, are also proposed in the end.

Building autonomous systems that achieve human level intelligence is one of the primary objectives in artificial intelligence (AI). It requires the study of a wide range of functions robustly across different phases of human cognition. This paper presents a review of agent cognitive architectures in the past 20 year's AI research. Different from software structures and simulation environments, most of the architectures concerned are established from mathematics and philosophy. They are categorized according to their knowledge processing patterns-symbolic, emergent or hybrid. All the relevant literature can be accessed publicly, particularly through the Internet. Available websites are also summarized for further reference.

A series of flow estimation problems, especially origin-destination estimation, involves optimally locating sensors on a transportation network to measure traffic counts. As compressed sensing (CS) provides a new method to solve the estimation problem, its sensor location strategy needs to be researched in order to facilitate the reconstruction. This paper first points out that the accurate flow recovery is difficult by introducing a necessary condition, and then categorizes the location determination into two cases: sensor number with restriction and without restriction. For both cases, we elucidate their theoretical foundations of locating methods and propose an algorithm based on column coherence minimization, which optimally facilitates the reconstruction for CS framework. Numerical experiments indicate that the selected sensor locations fit the flow recovery and the proposed algorithm, compared with other methods, can lead to a slightly better result under certain observations.

Generating travel behavior based on artificial population and an activity plan is a conventional method for traffic simulation. As a complicated and important constituent of travel behavior, destination selection is a decision-making process for space transfer and has been studied extensively in the disaggregate model. However, existing selection models only focus on the psychology or custom of individuals from a microscopic perspective and rarely take account of the actual traffic state. This causes a large deviation in simulation results and thus results in some obstacles for application. In this paper, a new destination selection model based on link flows is proposed. Further, a searching algorithm for an observed link set is given, and compressed sensing is used in the model solution. Experiments demonstrate that this model can predict the actual traffic state in rush hours quite well. Therefore, it contributes to the credible simulation and computational experiments.