Engineering topics
Publications and source records attributed to Greene, C. S..
Information on the identification and control of spacecraft is given. Maximum likelihood estimation, identification accuracy issues, steady state identifiability analysis and stochastic error with process noise are among the topics addressed.
A cost tradeoff is postulated for a stiff structure utilizing minimal controls (and control expense) to point and stabilize the vehicle. Extra costs for a stiff structure are caused by weight, packaging size, etc. Likewise, a more flexible vehicle should result in reduced structural costs but increased costs associated with additional control hardware and data processing required for vibration control of the structure. This tradeoff occurs as the ratio of the control bandwidth required for the mission to the lowest (significant) bending mode of the vehicle. The cost of controlling a spacecraft for a specific mission and the same basic configuration but varying the flexibility is established.
Qualitative and quantitative aspects of the multiple model adaptive control method are detailed. The method represents a cascade of something which resembles a maximum a posteriori probability identifier (basically a bank of Kalman filters) and a bank of linear quadratic regulators. Major qualitative properties of the MMAC method are examined and principle reasons for unacceptable behavior are explored.
The purpose of this paper is to summarize some results obtained for the adaptive control of the F-8C aircraft using the so-called MMAC method. The discussion includes the selection of the performance criteria for both the lateral and the longitudinal dynamics, the design of the Kalman filters for different flight conditions, the 'identification' aspects of the design using hypothesis testing ideas, and the performance of the closed-loop adaptive system.
The purpose of this paper is to summarize results obtained for the adaptive control of the F-8C aircraft using the so-called MMAC method. The discussion includes the selection of the performance criteria for both the lateral and the longitudinal dynamics, the design of the Kalman filters for different flight conditions, the 'identification' aspects of the design using hypothesis testing ideas, and the performance of the closed loop adaptive system.