BBIPS: a blackboard-based integrated process supervision

被引:2
作者
Oh, KW [1 ]
Quek, C [1 ]
机构
[1] Nanyang Technol Univ, Sch Comp Engn, Intelligent Syst Lab, Singapore 637989, Singapore
关键词
blackboard-based supervisory framework; BB-IPS; hierarchical structured; blackboard and regime scheduler; scheduling policed; control knowledge sources;
D O I
10.1016/S0952-1976(02)00012-X
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
As the controlled plant becomes more complex and dynamic, it usually requires more than one control technique to control the plant effectively. A blackboard framework allows different knowledge sources to corporate together to provide an opportunistic problem solving process than any one of the knowledge sources (e.g. fuzzy rule-base control. PID control. MRAC, etc.) can achieve alone. However. there is a general lack of structure in its domain knowledge sources. which often produces a complex control strategy. The integrated process supervision (IPS) framework divides the control regimes into three generic control grouping and uses a supervisory module to select the appropriate control regime to be activated according to the system performance. The IPS further emphasizes on the importance of having a good supervisory module. BBIPS integrates the blackboard and IPS frameworks and attempts to combine their respective merits. BBIPS is able to incorporate multiple and different control techniques within a common framework under the control of a supervisory module. In addition. it provides a hierarchically structured architecture that is both generic and modular in nature. which allows simple and easy addition of new knowledge sources. Extensive experimentations have been conducted using a three tanks flow system as a test-bed. This paper provides a detailed description of the architecture and the analysis of the empirical result. (C) 2002 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:703 / 714
页数:12
相关论文
共 19 条
[1]   Improved MCMAC with momentum, neighborhood, and averaged trapezoidal output [J].
Ang, KK ;
Quek, C .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2000, 30 (03) :491-500
[2]  
ARZEN KE, 1987, LUTFD2TFRT0291199 CO
[3]  
ASTROM K, 1983, IFAC WORKSH AD SYST
[4]  
Astrom K. J., 1989, IEEE Control Systems Magazine, V9, P60, DOI 10.1109/37.24813
[5]  
Astrom K. J., 1995, ADAPTIVE CONTROL
[6]  
Chen J., 1991, Proceedings. The Twenty-Third Southeastern Symposium on System Theory, P323, DOI 10.1109/SSST.1991.138571
[7]  
EMGELMORE R, 1988, BLACKBOARD SYSTEM
[8]   ARCHITECTURE FOR INTEGRATED PROCESS SUPERVISION [J].
LEITCH, R ;
QUEK, C .
IEE PROCEEDINGS-D CONTROL THEORY AND APPLICATIONS, 1992, 139 (03) :317-327
[9]  
LEITCH R, 1991, IEE P 3 INT C CONTR, V1, P127
[10]  
Lin C.T., 1996, NEURAL FUZZY SYSTEMS