What is an “intelligent paradigm”? While it is difficult to provide an exact definition for this term, the interests in intelligent paradigms research generally focus on designing and developing of computerized machines or systems that exhibit the capabilities of learning from experience, adapting to the surrounding environment, as well as understanding and controlling ones thinking or reasoning process. These are some crucial characteristics of intelligent paradigms so that they can be deployed as usable and useful tools to assist humans in daily activities. Indeed, recent research and development in intelligent paradigms has opened up the way for a number of theoretical advances and successful applications of intelligent techniques and approaches in various domains. Diverse intelligent paradigms are available in the literature, encompassing, to name a few, artificial neural networks, evolutionary algorithms, multiagent systems, artificial immune systems, swarm intelligence, knowledge-based systems, case-based reasoning, as well as hybrid intelligent systems in which these paradigms are contained. Applications of intelligent paradigms also span across various fields, covering, to name a few, information processing, decision making, control and robotics, industrial and medical diagnosis, data mining, e-learning and e-commerce, knowledge management, as well as virtual reality and multimedia. In this special issue, a total of eight articles are collected to showcase a small fraction of some recent advances in theory and application of intelligent paradigms. These articles present research into various intelligent paradigms, with their effectiveness in tackling different real-world problems demonstrated and discussed. A brief outline of each article is as follows. A powered wheelchair is an effective vehicle to help elderly or handicapped people move around the ordinary areas. Song et al. design of an electromyogram (EMG) pattern classifier that is robust against muscular fatigue effects for powered wheelchair control. It is discovered that variations of feature values owing to the effect of muscular fatigue are consistent for sustained duration. This finding leads to a new fatigue compensation method, and the fuzzy Min-Max neural network is employed as a robust EMG-based pattern classifier through the adaptation process of its hyperboxes. The proposed approach demonstrates improved performance for continuous control of powered wheelchair. Path planning is a fundamental problem in mobile robotics. The work by Chakraborty et al. addresses the issue of multi-robot path planning by using parallel differential evolution algorithms. Both centralized and distributed realizations for multi-robot path planning are studied, and the performances of the methods are compared with respect to a few pre-defined yardsticks. Most industrial plants are complex, nonlinear, time varying with time delay, and difficult to control. A fuzzy logic-based system is suitable for their control as it combines measurements, experts’ knowledge and op-
Read full abstract