提出了一种连接主义方法，利用移动机器人自身的时空经验，在缺乏全局坐标信息和环境先验模型的情况下，建立面向目标的认知地图。在线形成的时序处理网络（TSPN）可提供简洁的历史感知信息，以神经元激活特性保存空间知识，引导机器人运动。结合TSPN和反应式行为模块的导航系统可实现动态的路标及方向检测、路径学习和实时导航功能。仿真和实际实验验证了系统的有效性和适应性。

frame of the Intelligence Science has been proposed, the general features of the research objects of the Intelligence Science have been summarized, the relationships among the industrialization, informatization, intellectualization and automation have been investigated in this paper. The Intelligence Science consists of three portions: scientific foundation, technical methodology and application fields. The general features of Intelligence Science include complexity, intersection, nonlinearity, anthropomorphic property, uncertainty, incompleteness and distribution etc. The industrialization has been advanced in the direction: Primary Industrialization → Automatic Industrialization → Information Industrialization → Intelligent Industrialization. The proposed new scientific discipline would reflect the new height, new thought and new way for developing the control and automation science from one angle of view, and present a strong wish for establishing the new discipline of Intelligence Science.

In this paper, a frame of natural computation has been proposed in order to deal with the challenge from new computing braches such as DNA computation and quantum computation to traditional computing theory, and computation is considered as the inherent mechanism of the nature and the foundation of intelligence. Simulating computation has been created and improved viainspiration of natural computation, and Artificial Intelligence has been stablished on simulation of human brain computing pattern. According to this view, generalized mapping from computing panem existing in the nature to current burgeoning new computing panem inspired by the nature has been modeled, and then a coding and control mechanism of natural CO putation bas been analyzed with formalizing tools. So it can be concluded the theow foundation, intelligent power, and application hme will improve the sciences, such as computer science, artificial intelligence etc.

提出一种移动机器人运行环境的非光滑建模以及全局路径划方法，能够在离散的传感器信息基础上构造近似的Voronoi边界网络作为环境模型；采用基于最优竞争机制的进化计算与紧缩优化后，实现了移动机器人在复杂环境下的全局路径规划。

如何提高机器人的自主性和适应性是目前人工智能领域的热点问题之一。传统的集中式规划设计方法难以满足这些性能要求，将进化机制引入自适应机器人的设计为解决这一问题开辟了新的思路。系统地介绍了进化机器人学的研究背景、主要内容和现状，将其与传统人工智能方法和基于行为的机器人学进行比较，探讨了该领域研究中有待解决的问题,并对今后工作进行了展望。