Resource-Constrained Embedded Control and Computing Systems
Abstract: This thesis deals with methods for handling resource constraints in embedded control systems and real-time computing systems. By dynamic feedback-based resource scheduling it is possible to achieve adaptability andincreased performance for these systems.
A feedback scheduling strategy is presented, which uses feedback from plant states to distribute computing resources optimally among a set of controller tasks. Linear-quadratic controllers are analyzed, and expressions relating the expected cost to the sampling period and the plant state are derived and used for on-line sample-rate adjustments.
A flexible implementation of model predictive control (MPC) tasks is described. A termination criterion is derived that, unlike traditional MPC, takes the effects of computational delay into account in the optimization. A scheduling scheme is also described, where the MPC cost functions being minimized are used as dynamic task priorities for a set of MPC tasks.
A method for optimizing the use of computational resources in a multi-camera-based positioning system is studied. The covariance of the estimation error is minimized, while meeting computation time constraints.
A novel predictor for delay control in server systems is introduced. The predictor uses instantaneous measurements of queue length and arrival times and is continuously updated as new requests arrive according to a receding horizon principle. The predictor is evaluated in simulation and by experiments on an Apache web server.
The MATLAB/Simulink-based simulator TrueTime is presented. TrueTime is a codesign tool that facilitates simulation of distributed real-time control systems. TrueTime also supports simulation of wireless communication and resource constraints associated with wireless sensor/actuator networks.
CLICK HERE TO DOWNLOAD THE WHOLE DISSERTATION. (in PDF format)