RIA: Adaptive Neurocontrol Using Concepts from Immunology (REU Supplement)

本研究项目涉及自动控制的一个重要领域，即鲁棒自适应控制。主要目标是在免疫学概念和自适应神经控制概念之间建立一个强有力的类比，并利用这种类比来加强自适应神经控制的当前实践。免疫学是一门科学，研究所有生物体内存在的保护它们免受外部攻击的内在防御机制。免疫学与鲁棒适应性控制问题有许多相似之处。这个项目假设在神经控制、免疫学和遗传算法的概念之间存在匹配，如果这种匹配被正确理解，在自适应神经控制方面有很大的潜力可以扩大目前的趋势。本文的研究目标与这一匹配的实际实现有关。自适应神经控制如果得到充分发展，可以成功地应用于自主空间探索和机器人规划、高性能飞机/导弹控制、人机交互、国家航天飞机，柔性结构控制。建议使用上述两个领域来测试本提案中提出的概念。它们是：(a)人机交互问题中的自适应控制；(2)高性能飞机的鲁棒自适应控制。预计这些应用将揭示所提出的免疫自适应神经控制方法的基本特征。

This research project addresses an important area in automatic control, namely, robust adaptive control. The primary goal is to draw a strong analogy between concepts from immunology and concepts from adaptive neurocontrol and to use this analogy to enhance the current practices in adaptive neurocontrol. Immunology is the science of in-built defense mechanism that is present in all living beings to protect them against external attacks. The science of immunology has many parallels to the robust adaptive control problem. This project assumes that there is a match between the concepts of neurocontrol, immunology, and genetic alogortihms, and if this match is properly understood, there is a great potential for scaling up current trends in adaptive neurocontrol. Objectives of the research are related to the actual realization of this match. Adaptive neurocontrol, if fully developed, could be successfully used in autonomous space explorations and robot planning, high performance aircraft/missile control, man-machine interaction, the national aerospace plane, and flexible structure control. Two of the above areas are proposed to be used to test the concepts presented in this proposal. These are: (a) adaptive control in man-machine interaction problems; and (2) robust adaptive control of a high performance aircraft. It is anticipated that these applications will bring out the essential features of the proposed immunized adaptive neurocontrol approach.