Razali Samin

A New Path Estimation Strategy for Predicting Blind Persons' Motion in Indoor Environments

Abstract Research on the characteristics of spatial cognition without vision is used to improve the design of indoor environments to be safer for blind and visually handicapped persons. A fuzzy cognitive map (FCM) decision mechanism is presented for modeling path planning strategies adopted by blind travelers including wall-following, and shortcutting through the environment. A statistical case based reasoning (CBR) strategy is introduced for anticipating the points of switch between the two mentioned behaviours along the path. The combination of CBR and FCM modules provided a robust model of decision making which can be used for predicting blind motions. In this research, 51 eye-masked subjects contributed for obtaining the path patterns and for validating the results obtained using the proposed path prediction approach.

A hybrid genetic-heuristic algorithm for scheduling of automated guided vehicles and quay cranes in automated container terminals

Containers have been used in past decades increasingly as one of the most important transportation tools. Containers have revolutionized cargo shipping and thus changed the world trade systematically. Container terminals as the transhipment facility play a valuable role in performance of this transportation system. Improvement of this facility has been widely considered in literatures. Automated container terminals (ACTs) have been introduced to pursue this purpose. In ACTs various transport vehicles are automated and integrated to each other. Automated guided vehicles (AGVs) are used in ACTs to handle containers between quay cranes and storage yards. Usually scheduling of the AGVs is known as the key factor to improve the performance of ACTs. This paper proposed a heuristic algorithm to schedule the AGVs concurrently with quay cranes. A genetic algorithm is proposed to optimize the simultaneous scheduling of AGVs and QCs. The results showed that proposed genetic algorithm can be used in practical implications while its running time is reasonably low.

Artificial Neural Network (ANN) Approach for Predicting Friction Coefficient of Roller Burnishing AL6061

Artificial Neural Network (ANN) approach is a fascinating mathematical tool, which can be used to simulate a wide variety of complex scientific and engineering problems. Due to its highly reliable prediction quality, the usage of it is growing rigorously and had already become an ultimate tool for various applications in the field of engineering. In this study an ANN technique was used to predict friction coefficient of roller burnishing AL6061 for two orientations which is parallel burnishing orientation (PB) and cross burnishing orientation (CB). The input parameters were defined by widths of roller curvature (7.5mm, 8mm and 8.5mm), burnishing speeds (110rpm, 230rpm, 330rpm and 490rpm), and burnishing forces (155.06N, 197.45N, 239.83N and 282.22N) while the output parameter was friction coefficient. 173 data was used for training the ANN and another 115 data was used to test the ANN. 60 different configurations of ANN was trained by using 6 different training algorithms. It was found that feed-forward back-propagation network with 15 neurons in hidden layer that was trained by Levenberg-Marquardt training algorithm gave the best result when compared to other training algorithms used. From the results it was found that the training performance and prediction performance was 0.000809 and 0.710 respectively. From this study, it became obvious that the selected ANN with the configuration and training algorithm proved to be the most suitable among the other ANN investigated for similar applications.

A New Genetic-Fuzzy Algorithm for Mobile Robot Way-Finding in Environments with Any Types of Concave Obstacle

A new behavior-based algorithm is developed for reactive navigation of mobile robots. While fuzzy logic body of the algorithm performs the main tasks of obstacle avoidance and target seeking, an actual-virtual target switching is used to resolve the problem of limit cycles in any types of concave obstacles. The overall performance of the algorithm is then enhanced by using GA optimization of the functions. In this work, concave obstacles may have any shape such as corner, U-shape cul-de-sac, snail shape, or any other complicated shape. Trajectories and behavior analysis of a Pioneer robot are demonstrated to prove the robustness of the proposed algorithm.

Tremor suppression for 4-DOFs biodynamic hand model using genetic algorithm

A person who has severe hand tremor will have difficulty in doing specific tasks such as eating, combing or holding any objects. Currently, there is no medication that can cure the tremor. Thus, this study proposes the active tremor control, in which an intelligent controller is applied to suppress the hand tremor. The main objective is to optimise the proportional-integral (PI) controller using genetic algorithm (GA). A linear voice coil actuator (LVCA) is applied onto a four degree of freedom (4-DOF) human hand model represented in state space. The findings of the study demonstrate that the PI controller optimised by GA gives excellent performance in reducing the tremor error. Based on the frequency evaluation, the PI controller performance was roughly around 84% in reducing the peak of simulated hand tremor. The outcomes provide an important contribution towards achieving novel methods in suppressing hand tremor model by means of intelligent control.