Dan Henriksson

How does control timing affect performance? Analysis and simulation of timing using Jitterbug and TrueTime

To achieve good performance in systems with limited computer resources, the constraints of the implementation platform must be taken into account at design time. To facilitate this, software tools are needed to analyze and simulate how timing affects control performance. This article describes two such tools: Jitterbug and TrueTime.

TrueTime: Simulation of Control Loops Under Shared Computer Resources

Abstract The paper presents TrueTime, a Matlab/Simulink-based simulator for real-time control systems. TrueTime makes it possible to simulate the temporal behavior of multi-tasking real-time kernels containing controller tasks and to study the effects of CPU and network scheduling on control performance. The simulated real-time kernel is event-driven and can handle external interrupts as well as fine-grained details such as context switches. Arbitrary scheduling policies may be defined, and the control tasks may be implemented using C functions, M functions, or Simulink block diagrams. A number of examples that illustrate the use of TrueTime are presented.

Simulation of Wireless Networked Control Systems

Embedded systems are becoming increasingly networked and are deployed in application areas that require close interaction with their physical environment. Examples include distributed mobileagents and wireless sensor/ac tuator networks. The complexity of these applic ations make co- simulation a necessary tool during system development. This paper presents a simulation environment that facilitates simu- lation of computer nodes and communic ation networks inter- acting with the continuous-time dynamics of the real world. Features of the simulator include interrupt handling, task scheduling, wired and wireless communication, local clocks, dynamic voltage scaling, and battery-driven operation. Two simulation case studies are presented: a simple communication scenario and a mobile robot soccer game.

Optimal On-line Sampling Period Assignment for Real-Time Control Tasks Based on Plant State Information

The paper presents a feedback scheduling strategy for multiple control tasks that uses feedback from the plant states to distribute the computing resources optimally among the 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. In the case of minimum-variance control of multiple integrator processes, an exact expression for the optimal sampling periods is obtained. For the general case, an on-line optimization procedure is developed. The approach is exemplified on a set of controllers for first-order systems. The issues of computational delay and the choice of the feedback scheduler period are also discussed.

Improved prediction for Web server delay control

Control methods are being used increasingly for uncertainty management and QoS in modern Web server systems. Previous approaches have suggested combined feedforward and feedback control strategies, using queuing theory for feedforward delay prediction. While queuing theory allows one to predict delay as a function of arrival and service rates, the prediction applies only to long-term averages, and is therefore insensitive to sudden load changes. Unfortunately, Internet load is very bursty, leaving room for predictor improvement. The main contribution of this paper is an extension of the combined feedforward/feedback framework in which the queuing model is replaced with a predictor that instead uses instantaneous measurements to predict future delays. The proposed strategy is evaluated in simulation and by experiments on an Apache Web server. It is shown that the approach performs better than the combined queuing model based feedforward and feedback control presented in earlier papers.

Feedback scheduling of model predictive controllers

The paper presents some preliminary results on dynamic scheduling of model predictive controllers (MPCs). In an MPC, the control signal is obtained by on-line optimization of a cost function, and the MPC task may experience very large variations in execution time from sample to sample. Unique to this application, the cost function offers an explicit, on-line quality-of-service measure for the task. Based on this insight, a feedback scheduling strategy for multiple MPCs is proposed, where the scheduler allocates CPU time to the tasks according to the current values of the cost functions. Since the MPC algorithm is iterative, the feedback scheduler may also abort a task prematurely to avoid excessive input-output latency. A case study is presented, where the new approach is compared to conventional fixed-priority and earliest-deadline-first scheduling. General problems related to the real-time implementation of MPCs are also discussed.

Resource management for control tasks based on the transient dynamics of closed-loop systems

This paper presents a resource management strategy for control tasks that maximizes control performance within the available resources by readjusting the task periods at runtime. A feedback scheduler is used to determine on-line the optimal task periods considering the response over a finite time horizon of the plants controlled by arbitrary linear control laws. We show how this problem can be expressed as an optimization problem, where the objective function relates the sampling periods to the transient responses of the controlled plants, and where restrictions are based on EDF schedulability constraints. For the general case, the solution of the optimization problem is computationally expensive, and thus, an approximate procedure to be executed on-line has been developed. We present simulation results that validate the presented approach

On dynamic real-time scheduling of model predictive controllers

The paper discusses dynamic real-time scheduling in the context of model predictive control (MPC). Dynamic scheduling in this setting is motivated by the highly varying execution times associated with MPC controllers. Premature termination of the optimization algorithm is exploited to trade off prolonged computations versus computational delay. A feedback scheduling strategy for multiple MPC controllers is also proposed, where the scheduler allocates CPU time to the tasks according to the current values of the cost functions. Simulated examples show how the overall control performance may benefit from the application of the proposed schemes.

Conclusions of the ARTIST2 roadmap on control of computing systems

he use of control-based methods for resource manage- ment in real-time computing and communication systems has gained a substantial interest recently. Applications ar- eas include performance control of web-servers, dynamic resource management in embedded systems, traffic con- trol in communication networks, transaction management in database servers, error control in software systems, and au- tonomic computing. Within the European EU/IST FP6 Net- work of Exellence ARTIST2 on Embedded System Design a roadmap on Control of Real-Time Computing Systems has recently been completed. The focus of the roadmap is how flexibility, adaptivity, performance and robustness can be achieved in a real-time computing or communication system through the use of control theory. The item that is controlled is in most cases the allocation of computing and communication resources, e.g., the distribution or schedul- ing of CPU time among different competing tasks, jobs, re- quests, or transactions, or the communication resources in a network. Due to this, control of computing systems also goes under the name of feedback scheduling. The roadmap is divided into six research areas: con- trol of server systems, control of CPU resources, control of communication networks, error control of software systems, feedback scheduling of control systems, and control mid- dleware. For each area an overview is given and challenges for future research are stated. The aim of this position paper is to summarize the conclusions concerning these research challenges. (Less)

Control loop timing analysis using truetime and jitterbug

A modern control system is typically implemented as a multitasking software application executing in a real-time operating system. If the computer load is high, the controller will experience delays and jitter, which in turn degrade the control performance. Arguing for an integrated design approach, the paper describes two computer tools for implementation-aware control analysis: TrueTime and Jitterbug. An example is given where the tools are used together to evaluate the performance of various control task implementations