Volume 26, Issue 4

Traffic incidents in big cities are increasing alongside economic growth, causing traffic delays and deteriorating road safety conditions. Thus, developing a universal freeway automatic incident detection (AID) algorithm is a task that took the interest of researchers. This paper presents a novel automatic traffic incident detection method based on the extreme learning machine (ELM) algorithm. Furthermore, transfer learning has recently gained popularity as it can successfully generalise information across multiple tasks. This paper aimed to develop a new approach for the traffic domain-based domain adaptation. The ELM was used as a classifier for detection, and target domain adaptation transfer ELM (TELM-TDA) was used as a tool to transfer knowledge between environments to benefit from past experiences. The detection performance was evaluated by common criteria including detection rate, false alarm rate, and others. To prove the efficiency of the proposed method, a comparison was first made between back-propagation neural network and ELM; then, another comparison was made between ELM and TELM-TDA.

Analyzing the learning dynamics in multi-agent systems (MASs) has received growing attention in recent years. Theoretical analysis of the dynamics was only possible in simple domains and simple algorithms. When one or more of these restrictions do not apply, theoretical analysis becomes prohibitively difficult, and researchers rely on experimental analysis instead. In experimental analysis, researchers have used some global performance metric(s) as a rough approximation to the internal dynamics of the adaptive MAS. For example, if the overall payoff improved over time and eventually appeared to stabilize, then the learning dynamics were assumed to be stable as well. In this paper, we promote a middle ground between the thorough theoretical analysis and the high-level experimental analysis. We introduce the concept of mining dynamics and propose data-mining-based methodologies to analyze multi-agent learning dynamics. Using our methodologies, researchers can identify clusters of learning parameter values that lead to similar performance, and discover frequent sequences in agent dynamics. We verify the potential of our approach using the well-known iterated prisoner’s dilemma (with multiple states) domain.

Currently, most artificial neural networks (ANNs) represent relations, such as back-propagation neural network, in the manner of functional approximation. This kind of ANN is good at representing the numeric relations or ratios between things. However, for representing logical relations, these ANNs have disadvantages because their representation is in the form of ratio. Therefore, to represent logical relations directly, we propose a novel ANN model called probabilistic logical dynamical neural network (PLDNN). Inhibitory links are introduced to connect exciting links rather than neurons so as to inhibit the connected exciting links conditionally to make them represent logical relations correctly. The probabilities are assigned to the weights of links to indicate the belief degree in logical relations under uncertain situations. Moreover, the network structure of PLDNN is less limited in topology than traditional ANNs, and it is dynamically built completely according to the data to make it adaptive. PLDNN uses both the weights of links and the interconnection structure to memorize more information. The model could be applied to represent logical relations as the complement to numeric ANNs.

Rainfall, being one of the most important components of the hydrological cycle, plays an extremely important role in agriculture-based economies like India. This paper presents a comparison between three soft computing techniques, namely Bayesian regression (BR), support vector regression (SVR), and wavelet regression (WR), for monthly rainfall forecast in Assam, India. A WR model is a combination of discrete wavelet transform and linear regression. Monthly rainfall data for 102 years from 1901 to 2002 at 21 stations were used for this study. The performances of different models were evaluated based on the mean absolute error, root mean square error, correlation coefficient, and Nash-Sutcliffe efficiency coefficient. Based on model statistics, WR was found to be the most accurate followed by SVR and BR. The efficiencies for the BR, SVR, and WR models were found to be 32.8%, 52.9%, and 64.03%, respectively. From the spatial analysis of model performances, it was found that the models performed best for the upper Assam region followed by lower, southern, and middle regions, respectively.

A large number of changeovers exist in multi-item and small lot size production mode, which lead to a long setup time and significantly influence productivity. However, setup time has received significantly less attention in previous studies regarding the single machine scheduling (SMS) problem. This paper investigates the SMS problem with sequence-dependent setup time (SDST). First, a model of SMS-SDST is developed to minimise the makespan. An earliest due date (EDD), SDST, and ant colony optimisation (ACO) heuristic algorithm is presented. In order to evaluate the performance of the proposed algorithm, Taguchi’s robust design method is employed to define the best parameter values. Related results are analysed by statistical tools. The computational experiments are compared with an ACO algorithm and a genetic algorithm. The experimental results and statistical analyses show that the proposed EDD-SDST-ACO is effective for the problem.

As the actual links in mobile ad hoc networks (MANETs) are dynamic, the feasible path generated may not exist during the transmission of multimedia data. There is a need for generating multiple paths for guaranteed transmission. The multi-path finding can be achieved through a genetic-based algorithm that faces the major challenge in quality-of-service (QoS)-aware routing. To enhance the QoS communication over the MANETs, an exponential genetic algorithm (GA)-based stable and load-aware QoS routing protocol (SLAQR) is proposed in this paper. This paper chiefly focuses on the enhancement of the GA-based routing algorithm by including exponential function in the searching process and modifying the fitness function, which incorporates QoS metrics like the node’s static resource capacity, dynamic resource availability, neighborhood quality, and link quality. The originality of the proposed protocol comes from the fact that it introduces multiple parameters into the route quality computation and integrates the exponential function into the GA. For experimental validation, the simulation of the proposed method is done and the results are compared with existing protocols such as GAQR (GA-Based QoS Routing), QMRB-AODV (QoS Mobile Routing Backbone Over Ad Hoc On-Demand Distance Vector Routing), and EISGA (Ensemble of Immigrant Strategies with Genetic Algorithm). From the outcome, we conclude that the delivery ratio, throughput, and delay of our proposed SLAQR approach have improved the GAQR approach by 58%, 56%, and 97%, respectively.

Manufacturing resources are shared and centrally managed on the cloud platform in cloud manufacturing, which is a new model of modern manufacturing. The production data are collected, which can be used to predict the manufacturing events. Based on those, disruption problems of scheduling should be researched from a new point of view. In this paper, new job arrivals were considered as the disruption event. The time of the occurrence of disruption was predictable in contrast to uncertainty. Alternative subcontractors chosen from the cloud platform were available for outsourcing with different processing prices and transporting distances. The objective of the original scheduling, the deviation between the new schedule and the old one, and the outsourcing cost were all considered. To express the problem, mathematical models and a three-field notation model were constructed. To solve the problem, a hybrid quantum-inspired chaotic group leader optimization algorithm was proposed, in which a hybrid encoding way was applied. To verify the algorithm, experiments were carried out. The results showed that the proposed algorithm performs well.

This paper presents a novel adaptive fuzzy high-order super-twisting sliding mode controller, based on the modified super-twisting control (STC), to achieve accurate trajectory tracking for a robotic manipulator with unknown structured uncertainties, parametric uncertainties, and time-varying external disturbances. Initially, a non-linear homogeneous sliding manifold is designed to achieve finite-time convergence, better robustness, and good transient characteristics. Afterwards, conventional STC is modified with the new sliding surface that eliminates the limitation of STC application only on relative degree 1 systems. Moreover, two adaptive fuzzy systems are designed to replace the STC signals for handling the chattering problem and overestimating the controller gains. These fuzzy systems are continuously adjusted by two adaptation laws that are deduced from the Lyapunov stability theory. These adaptive laws need only a sliding surface variable as an input and generate the optimal controller gains as an output. The finite-time convergence and stability of the proposed controller is analyzed by the homogeneous Lyapunov stability theory. Finally, to show the efficacy of the proposed method, the controller is simulated on a 2-degree-of-freedom planar robotic manipulator to obtain the accurate trajectory tracking. Simulation results demonstrate the superiority of the proposed control scheme in the presence of structured and unstructured uncertainties.

The importance of intelligent tutoring systems has rapidly increased in past decades. There has been an exponential growth in the number of ends users that can be addressed as well as in technological development of the environments, which makes it more sophisticated and easily implementable. In the introduction, the paper offers a brief overview of intelligent tutoring systems. It then focuses on two types that have been designed for education of students in the tertiary sector. The systems use elements of adaptivity in order to accommodate as many users as possible. They serve both as a support of presence lessons and, primarily, as the main educational environment for students in the distance form of studies – e-learning. The systems are described from the point of view of their functionalities and typical features that show their differences. The authors conclude with an attempt to choose the best features of each system, which would lead to creation of an even more sophisticated intelligent tutoring system for e-learning.

Artificial bee colony (ABC) is a kind of a metaheuristic population-based algorithms proposed in 2005. Due to its simple parameters and flexibility, the ABC algorithm is applied to engineering problems, algebra problems, and so on. However, its premature convergence and slow convergence speed are inherent shortcomings. Aiming at the shortcomings, a novel global ABC algorithm with self-perturbing (IGABC) is proposed in this paper. On the basis of the original search equation, IGABC adopts a novel self-adaptive search equation, introducing the guidance of the global optimal solution. The search method improves the convergence precision and the global search capacity. An excellent leader can lead the whole team to obtain more success. In order to obtain a better “leader,” IGABC proposes a novel method with global self-perturbing. To avoid falling into the local optimum, this paper designed a new mutation strategy that simulates the natural phenomenon of sick fish being eaten.