Timing Problems in Real-Time Control Systems

The industrial trend in embedded systems is than an increasing amount of control applications are implemented in distributed computer systems. Additional problems then face the designers such as significant and potentially varying end-to-end time-delays, the fact that data may be lost on the network, and that one or more nodes, or even parts of a node, may experience errors. Of course, similar effects can occur in single processor systems with run-time scheduling, but the problems become more apparent in distributed systems, which include aspects on scheduling processors and communications, node synchronization, and consistency of state variables. Without global synchronization, additional delays will arise in feedback loops that are closed over a network and in systems with several cooperating controllers working with different sampling periods (multirate systems). Many trade-offs exist depending on the requirements and the particular application considered. Research on timing problems in control systems was initiated early in the DICOSMOS project, Wittenmark et al. (1995). It turned out that very little work have been devoted to this interdisciplinary problem, neither from the control nor the computer engineering perspective. Different design constraints may necessitate a solution taken from both of these approaches. Interdisciplinary design and research approaches are obviously beneficial in that they open up new opportunities for designers as well as research directions. Within DICOSMOS we have attempted to tackle both these perspectives by following research directions on