Search

In some artificial systems, performing the realistic perception for actors, which will handle both the processes to simulate the sensing organs and identify, will spend most computational time. Unfortunately, this matter even ruins the result of decision based on perception. In order to reduce the computation cost from a systemic view and optimize the performance of system, a brand-new perceptual focuser was proposed. The perceptual focuser is the core of artificial animal sensation system, which provides the external environment and internal condition information for the behavior decision system. Artificial animal behavior memory formation is also the result of the focuser's analysis focusing. This paper proposes and analyzes two kinds of memory-form algorithm. The quadratic method to food spot which gains has made the variance computation, rejects noise spots which are apart from other normal spots, so that it obtains the expected food distribution position. In view of the quadratic method's insufficiency, the improvement mean-cluster algorithm profits from the data mining theory, which makes the noise-spotrejection accurately. Select the algorithm according to the different situation, which can make the focuser achieve the validity of the realization of artificial animal food memory. [ABSTRACT FROM AUTHOR]

The task-oriented proactive seamless migration is one of difficult problems to be solved in pervasive computing paradigm. Apparently, this function of seamless mobility is suitable for mobile services, such as mobile Web-based learning. But when seamless migration for computing task of learning is realized among PC, laptop, or PDA, there are several difficult problems to be solved, such as how to supply the proactive/attentive service with uncertainty for aware context. In order to realize E-learning based on proactive seamless migration, we design and improve relative fuzzy-neural approach (of course, besides it, there are other approaches). Generally, the network can be classified into two. One is that fuzzy logic reasoning is completed by fuzzy weight in neural system. The other is that the input data must be fuzzified in the first or second level, but not weight. We discuss and study the second in this paper. For proactive decision, fusion method based on fuzzy-neural can make Web-based learning system keep advantage of fuzzy logic system and remain adaptive optimum in proactive/attentive service. The correctness and validity of our new approach have been tested. [ABSTRACT FROM AUTHOR]