Collaborative Research: A Unified Cellular Self-Organizing Approach to Design Automation and Operation of Complex Systems

协作研究：复杂系统设计自动化和操作的统一蜂窝自组织方法

• 批准号: 1201107
• 负责人: Yan Jin
• 金额: $ 24万
• 依托单位: University of Southern California
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-06-01 至 2015-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1201107&HistoricalAwards=false
• 关键词: Collaborative; Research; Unified; Cellular; Self

This objective of this research award is to create a new generalization of cellular self-organization as a way to automate the design and control of complex and interdisciplinary systems. Concepts from nature such as evolution, cellular organization, and self-organizing behavior serve as the inspiration for the research. The researchers are creating a software framework built on four elements: graphs, fields, rules, and rule decisions. The graphs are used to define system architecture and dimensions and exist within a three-dimensional physics-based simulator environment with pertinent internal state variables (positions, velocities, accelerations, temperatures, stresses, etc.). Rules and rule decisions govern how graphs change in response to state variables and external field effects like gravity or temperature in a manner that mimics true physical and biological effects. The computational framework will enable the ?cyber-physical boundary? to be easily shifted. For example, the framework would serve as a design automation tool to design a device before fabrication (simulating the environment of the device virtually) and also used as a control system for the device in actual operation so that the device might adapt to changes in the environment. This would occur by monitoring real-world fields from external sensors and transforming the graph or configuration of the resulting reconfigurable device. The research is a generalization of various published techniques (e.g. cellular automaton, topology optimization, and claytronics) under a single method. Such a framework is necessary for solving interdisciplinary and multi-physics design problems such as those found in mechatronics and robotics. The results will be widely disseminated for greatest impact.

该研究奖的这个目标是创建对蜂窝自我组织的新概括，以此来自动化复杂和跨学科系统的设计和控制。来自自然的概念，例如进化，细胞组织和自组织行为，是研究的灵感。研究人员正在创建一个基于四个要素的软件框架：图，字段，规则和规则决策。这些图用于定义系统体系结构和尺寸，并存在于具有相关内部状态变量（位置，速度，加速度，温度，压力等）的三维物理模拟器环境中。规则和规则决策控制了图形如何响应状态变量和外部田间影响（例如重力或温度）的变化，以模仿真正的物理和生物学效应。计算框架将实现网络物理边界？容易移动。例如，该框架将作为设计自动化工具，用于在制造之前设计设备（实际上模拟设备的环境），并用作实际操作中设备的控制系统，以便设备可以适应环境的变化。这将通过从外部传感器监视现实世界字段并转换所得可重构设备的图形或配置来实现这一点。这项研究是对单个方法中各种已发表技术（例如细胞自动机，拓扑优化和克莱特龙）的概括。这样的框架对于解决跨学科和多物理设计问题（例如在机器人和机器人技术中发现的问题）是必要的。结果将被广泛传播，以产生最大的影响。