为了更好地满足复杂现实工程应用中问题的建模与求解需求，本项目致力于系统而深入地研究复杂未来时变问题，设计开发出用于求解该类问题的新型递归神经动力学模型，以同时满足附加多重约束和高精度实时计算的需求，并将所提出的模型应用于不同类型的冗余度机械臂以进行数值/实物实验及实践推广。另外，通过优化神经动力学设计法则，本项目也将进一步设计开发出新型递归神经动力学扰动抑制模型，使得其本身对在时变噪声干扰环境下的复杂未来时变问题高精度实时求解的同时也具备有效抑制时变噪声干扰的优越性能。本项目力争提出一套面向复杂未来时变问题高精度实时求解、附加约束有效处理、时变噪声有效抑制以及作为冗余度机械臂应用基础的科学理论，并设计开发出相应的离散模型/算法，且进行计算机数值验证、实物实验以及实践推广等。本项目的研究成果将有助于突破受低精度、弱实时性、附加约束限制以及时变噪声干扰的机械臂研发和实践中的核心关键技术。

To better meet the demands of problem modeling and solving in the complex practical engineering applications, this project is devoted to the systematical and in-depth research of complex future time-variant problems. The new recurrent neurodynamic models are designed and developed for solving such problems, so as to satisfy the requirements of additional multiple constraints and high-precision real-time computing. Then, the proposed models are applied to the numerical/physical experiments and practices for different types of redundant robot manipulators. In addition, by optimizing the design rules of neurodynamics, this project will further design and develop the new recurrent neurodynamic disturbance suppression models, so that they can effectively suppress time-variant noise perturbation when solving complex future time-variant problems with high precision in real time. This project aims to put forward a set of complex future time-variant problems oriented scientific theories with the high-precision real-time solutions, the effective treatment of additional constraints, the effective suppression of time-variant noises, and the application foundation of redundant robot manipulators, design and develop the corresponding discrete models/algorithms, and conduct computer numerical validations, physical experiments and practices, etc. The research results of this project will contribute to breaking through the core key technologies in the research, development, and practice of robot manipulators, which are limited by low precision, weak real-time performance, additional constraints, and time-variant noise disturbance.