SCHEDULE OPTIMIZATION FOR TTETHERNET-BASED TIME-TRIGGERED AUTOMOTIVE SYSTEMS |
Rui Zhao, Guihe Qin, Jie Yan, Jun Qin |
Jilin University |
|
|
|
|
ABSTRACT |
TTEthernet is a new high availability network protocol for the automotive domain, providing support for the transmissions of time-critical periodic frames in a time-triggered (TT) traffic and event-triggered frames in a rate-constrained (RC) or best-effort (BE) traffic. The design of a system scheduling with communication over the TTEthernet is not an easy task because of protocol and timing constraints as well as the demand for high quality-of-service and extensibility. In this paper, we study the co-optimization problem of the tasks and signals scheduling for TTEthernet-based time-triggered automotive systems. We propose a Mixed-Integer Linear Programming (MILP) framework for optimizing the signal packing, signal scheduling and task scheduling, subject to all protocol, processor and timing constraints with respect to latency- or extensibility- related metrics. Extensive experiments are conducted to evaluate the effectiveness and efficiency of the proposed approach in a variety of scenarios. |
Key Words:
Automotive, Time-triggered systems, TTEthernet, Scheduling, Mixed-integer linear programming |
|
|
|