针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息...针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息熵的概念定义了状态不确定性测度,据此定义了Q学习动作偏向信息函数,通过对Q学习奖励函数采用启发式回报函数设计,将动作偏向信息利用附加回报的方式融入学习系统,并证明了算法的收敛性和最优策略不变性.在学习过程中,Q学习根据偏向信息调整搜索空间,减少了Q学习必须探索的有效状态-动作对数目,同时偏向信息根据Q学习结果不断进行调整,避免了不正确的误导.经仿真实验比较,结果表明,该策略具有对动态环境的适应性和大状态空间下收敛的快速性,提高了调度效率.展开更多
Future manufacturing systems need to cope with frequent changes and disturbances, therefore their control architectures require constant adaptability, agility, stability, self-organization, intelligence, and robustnes...Future manufacturing systems need to cope with frequent changes and disturbances, therefore their control architectures require constant adaptability, agility, stability, self-organization, intelligence, and robustness. Bio-inspired manufacturing system can well satisfy these requirements. For this purpose, by referencing the biological organization structure and the mechanism, a bio-inspired manufacturing cell is presented from a novel view, and then a bio-inspired self-adaptive manufacturing model is established based on the ultra-short feedback mechanism of the neuro-endocrine system. A hio-inspired self-adaptive manufacturing system coordinated model is also established based on the neuro-endocrine-immunity system （NEIS）. Finally, an example based on pheromone communication mechanism indicates that the robustness of the whole manufacturing system is improved by bio-inspired technologies.展开更多
The welding fixtures are the most important devices for an auto body welding assembly line. The current special fixtures used by many automotive manufactures are only fit for one or several specific welding processes,...The welding fixtures are the most important devices for an auto body welding assembly line. The current special fixtures used by many automotive manufactures are only fit for one or several specific welding processes, and the dimensional problem in the circle due to several variation sources accumulation has no adjustment. The active error compensating welding fixture system for auto body is designed and manufactured. The detecting model, coordinate transformation model, and adjusting model based on auto body coordinate system are presented. The dowel pin modular design is adopted in the structure of the fixture to suit different workpieces with some similar characteristics. The online detection and adaptive control system using eddy current sensors and adaptive adjusting devices is analyzed. Three kinds of the left rear wheel covers SGM60 are selected to test workpieces of the developed system, and the active error compensating experiments are performed in the lab for many times. Test results show the validity of mechanism reconfigurations, on-line detections and error compensations of the developed welding fixture.展开更多
文摘针对知识化制造环境下的自适应调度问题,提出基于状态-动作不确定性偏向Q学习(state-action uncertainty bias based Q-learning,简称SAUBQ学习)的知识化制造自适应调度策略.该策略针对传统Q学习收敛速度慢,训练时间长等问题,引入信息熵的概念定义了状态不确定性测度,据此定义了Q学习动作偏向信息函数,通过对Q学习奖励函数采用启发式回报函数设计,将动作偏向信息利用附加回报的方式融入学习系统,并证明了算法的收敛性和最优策略不变性.在学习过程中,Q学习根据偏向信息调整搜索空间,减少了Q学习必须探索的有效状态-动作对数目,同时偏向信息根据Q学习结果不断进行调整,避免了不正确的误导.经仿真实验比较,结果表明,该策略具有对动态环境的适应性和大状态空间下收敛的快速性,提高了调度效率.
基金Supported by the National Natural Science Foundation of China (50505017)Fok Ying Tung Edu-cation Foundation (111056)+1 种基金the Innovative and Excellent Foundation for Doctoral Dissertation of Nanjing University of Aeronautics and Astronautics (BCXJ08-07)the New Century Excellent Talents in University,China (NCET-08)~~
文摘Future manufacturing systems need to cope with frequent changes and disturbances, therefore their control architectures require constant adaptability, agility, stability, self-organization, intelligence, and robustness. Bio-inspired manufacturing system can well satisfy these requirements. For this purpose, by referencing the biological organization structure and the mechanism, a bio-inspired manufacturing cell is presented from a novel view, and then a bio-inspired self-adaptive manufacturing model is established based on the ultra-short feedback mechanism of the neuro-endocrine system. A hio-inspired self-adaptive manufacturing system coordinated model is also established based on the neuro-endocrine-immunity system （NEIS）. Finally, an example based on pheromone communication mechanism indicates that the robustness of the whole manufacturing system is improved by bio-inspired technologies.
基金Shanghai Leading Academic Discipline Project,China(No.B602)Patent Second Development Project of Science and Technology Commission of Shanghai Municipality,China(No.05dz52038)
文摘The welding fixtures are the most important devices for an auto body welding assembly line. The current special fixtures used by many automotive manufactures are only fit for one or several specific welding processes, and the dimensional problem in the circle due to several variation sources accumulation has no adjustment. The active error compensating welding fixture system for auto body is designed and manufactured. The detecting model, coordinate transformation model, and adjusting model based on auto body coordinate system are presented. The dowel pin modular design is adopted in the structure of the fixture to suit different workpieces with some similar characteristics. The online detection and adaptive control system using eddy current sensors and adaptive adjusting devices is analyzed. Three kinds of the left rear wheel covers SGM60 are selected to test workpieces of the developed system, and the active error compensating experiments are performed in the lab for many times. Test results show the validity of mechanism reconfigurations, on-line detections and error compensations of the developed welding fixture.
