De-Yuan Huang

A Simple Approach to Implementing a System for Monitoring Driver Inattention

Over the past few years there has been an increased interest in developing systems to help drivers drive safely. Although driver fatigue is an important factor in a large number of traffic accidents, a certain percentage of accidents can be due to the inattention factor. In this paper, a low-cost driver inattention monitoring system is proposed. This system can not only detect driver fatigue but also detect whether the driver is looking at other directions for an extended period of time. The system alerts the driver when one of the above two driving conditions happens.

Mapping multi-spectral remote sensing images using rule extraction approach

To improve the accurate rate of mapping multi-spectral remote sensing images, in this paper we construct a class of HyperRectangular Composite Neural Networks (HRCNNs), integrating the paradigms of neural networks with the rule-based approach. The supervised decision-directed learning (SDDL) algorithm is also adopted to construct a two-layer network in a sequential manner by adding hidden nodes as needed. Thus, the classification knowledge embedded in the numerical weights of trained HRCNNs can be extracted and represented in the form of If-Then rules. The rules facilitate justification on the responses to increase accuracy of the classification. A sample of remote sensing image containing forest land, river, dam area, and built-up land is used to examine the proposed approach. The accurate recognition rate reaching over 99% demonstrates that the proposed approach is capable of dealing with image mapping.

A reinforcement-learning approach to robot navigation

This paper presents a reinforcement-learning approach to a navigation system which allows a goal-directed mobile robot to incrementally adapt to an unknown environment. Fuzzy rules which map current sensory inputs to appropriate actions are built through the reinforcement learning. Simulation results illustrate the performance of the proposed navigation system. In this paper, ACSNFIS is used as the main network architecture to implement the reinforcement-learning based navigation system.

Application of a SOM-Based Optimization Algorithm in Minimizing Construction Time for Secant Pile Wall

Construction time matters for activities where rental equipment must be used. The building of a secant pile wall requires the rental of equipment and finding the optimal sequence to minimize the construction time is one way to lower construction costs. In this study we develop an effective and efficient optimization algorithm, which we call self-organizing feature map (SOM)-based optimization (SOMO), to minimize the construction time. The algorithm is applied to a case study to obtain the optimal sequences for both primary and secondary bored piles for a secant pile wall. The new SOMO algorithm is developed based on the ability of the human brain to produce topologically ordered mapping, so as to exploit better solutions via updating the weighting vectors of the neurons in a self-organizing topological way that occurs in the evolution of the feature map for optimization. Given detailed building time of the 16 activities of each bored pile, we find that 143.92 h or 27.21% of the original construction can be saved. The optimal sequences for both primary and secondary bored piles are also determined. The practicability of the SOMO algorithm is substantiated.

A PSO-based Decision Aid for Multi-Aircraft Combat Situations

In this paper, the problem of assigning aircraft to protect assets from attack by enemy aircraft is formulated as a constrained resource allocation problem. The aircraft assignment problem is an NP-complete problem. A PSO-based approach incorporating four fuzzy measures of the predominance values for solving the problem is developed to help a military planner reduce the required planning time and provide a better strategy. Several simulations are used to illustrate the effectiveness of the proposed decision aid.

SOM-Optimized Neurofuzzy Classifiers for Measuring Expatriation Willingness

The authors developed a self-organizing map-based optimization (SOMO) neurofuzzy classifier to deal with a practical expatriation willingness (EW) problem, which is associated with employees’ willingness to accept expatriate assignments. The proposed model also delivers more information about rule coverage and generates user-friendly outcomes. The authors adopt the SOMO algorithm to optimize the weights of neurons in the proposed hybrid neurofuzzy classifier. They evaluated the model’s feasibility using the databases from previous studies exploring employees’ EW. The results show that the proposed neurofuzzy classifier yields 11 determination rules whose accuracy rates are greater than 80 percent. The contribution to the body of knowledge lies in the significant improvement in accuracy rates and coverage for predicting EW rules, optimization for all the network parameters using a novel algorithm, and more friendly outputs for practical use.

Application of a Learning-Companion Robot in Learning Environments

In this paper, a learning-companion robot based on a Pleo (a dinosaur-robot) is introduced. The learning-companion robot is equipped with a low-cost Web camera, a computer, and an attention monitoring algorithm so that it can not only monitor a learnerpsilas attention status but also detect whether the learner is too close to a computer screen during an on-line learning activity. The motivation of the development of the learning-companion robot is to alleviate parentspsila supervision burdens when they can not personally accompany their children to engage in some learning activities or playing computer games.