Kernel-based architectures have been proposed as a possible solution to build safe cooperative systems with improved performance. These systems adjust their operation mode at run-time, depending on the actual quality of sensor data used in control loops and on the execution timeliness of relevant control functions. Sets of safety rules, defined at design-time, express the conditions concerning data quality and timeliness that need to be satisfied for the system to operate safely in each operation mode. In this paper we propose a solution for practically expressing these safety rules at design-time, and for evaluating them at run-time. This evaluation is done using periodically collected information about safety-related variables. For expressing the rules we adopt the XML language. The run-time solution is based on a safety rules evaluation engine, which was designed for efficiency and scalability. We describe the architecture of the engine, the solution for structuring data in memory and the rule evaluation algorithm. A simple sensor-based control system is considered to exemplify how the safety rules are expressed.
