Integrated decision and control of human-engineered complex systems

Abstract

This paper presents a comprehensive decision and control strategy for human-engineered complex systems to achieve simultaneously the following objectives: (i) high-performance with quality assurance; (ii) reliability and structural durability with extended service life and (iii) operability over a wide range. Results from several systems-theoretic disciplines, such as probabilistic robust control (PRC), damage mitigating control (DMC), health and usage monitoring (HUM) and discrete event supervisory (DES) decision and control have been synergistically combined to achieve the above goal. The proposed decision and control system is hierarchically structured with two-tier architecture. The lower tier incorporates continuously-varying control that is designed using a combination of PRC and DMC, and the upper tier is designed to provide information and intelligence through DES decision and control that monitors the system response for detection and mitigation of anomalous behaviour, performance degradation and potential degradation of structural durability. To assure desired quality at permissible levels of risk as well as under different operating conditions, the PRC at the lower tier makes a trade off between robustness and performance, while damage mitigation in critical structures is achieved via DMC that also facilitates health and usage monitoring of the complex system. Based on the information derived from the observed time series data, the DES decision and control at the upper tier may decide to switch, in real time, to one control module from another in order to satisfy the specified performance and safety requirements. The switching actions are executed at the lower tier. The integrated system, including the proposed decision and control architecture, has been tested and validated on a rotorcraft simulation test bed.

Keywords:

• Probabilistic robust control
• Damage mitigating control
• Health and usage monitoring
• Discrete event supervisory control

Acknowledgements

The authors would like to thank Yasar Murat, Jialing Chen and Derek Bridges for their support in constructing and maintaining the rotorcraft simulation test bed on which the various simulation experiments were performed. The work was supported in part by the US Army Research Laboratory and the US Army Research Office under Grant No. DAAD19-01-1-0646 and NASA Glenn Research Center under Grant No. NNC04GA49G.