Data-Age Analysis and Optimisation for Cause-Effect Chains in Automotive Control Systems

Automotive control systems typically have latency requirements for certain cause-effect chains. When implementing and integrating these systems, these latency requirements must be guaranteed e.g. by applying a worst-case analysis that takes all indeterminism and limited predictability of the timing behaviour into account. In this paper, we address the latency analysis for multi-rate distributed cause-effect chains considering static-priority preemptive scheduling of offset-synchronised periodic tasks. We particularly focus on data age as one representative of the two most common latency semantics. Our main contribution is an Mixed Integer Linear Program-based optimisation to select design parameters (priorities, task-to-processor mapping, offsets) that minimise the data age. In our experimental evaluation, we apply our method to two real-world automotive use cases.

I. Introduction

Automotive control systems are a prominent representative of increasingly complex Cyber-Physical Systems (CPSs), particularly with regard to the emergence of Advanced Driver-Assistance Systems (ADASs). These systems are typically constrained by certain Sensor-to-Actuator (StA) delays that must not be exceeded. These delays are commonly captured as (worst-case) latencies of so-called cause-effect chains, which describe the data flow as it is being processed by different parts (i.e. tasks) of the software system. In particular, when it comes to multi-rate control systems, sampling of data values is applied. This leads to a time-triggered (periodic) data exchange along the cause-effect chains which is well-understood by control theory. Note that, depending on the function, i.e. the particular sensor and actuator, one distinguishes between two latency semantics: reaction time and data age (cf. [[1]]). The former describes the latest time it takes a control system to react to a certain sensor value whereas the latter describes the maximum age of the input data that was used for a particular output. A too large data age often reduces the control performance such that data age constraints must be considered for certain chains.