Learning-Based Response Time Analysis in Real-Time Embedded Systems: A Simulation-Based Approach

Response time analysis is an essential task to verify the behavior of real-time systems. Several response time analysis methods have been proposed to address this challenge, particularly for real-time systems with different levels of complexity. Static analysis is a popular approach in this context, but its practical applicability is limited due to the high complexity of the industrial real-time systems, as well as many unpredictable runtime events in these systems. In this work-in-progress paper, we propose a simulation-based response time analysis approach using reinforcement learning to find the execution scenarios leading to the worst-case response time. The approach learns how to provide a practical estimation of the worst-case response time through simulating the program without performing static analysis. Our initial study suggests that the proposed approach could be applicable in the simulation environments of the industrial real-time control systems to provide a practical estimation of the execution scenarios leading to the worst-case response time.

1 Introduction

Nowadays, embedded systems have become important parts of our life and are becoming more complex and powerful. Real-time programs running on embedded systems are key parts of many industrial control systems like those in telecommunication, railway, avionics, automotive, and medical care. Many of the industrial realtime embedded systems are referred to as complex real-time systems due to the internal functional complexity resulting from dependencies between tasks, asynchronous message-passing, runtime changeable priorities, and task offsets [1].