A Time-Sensitive Networking Traffic Scheduling Method Based on Q-Learning Routing Optimization

With the rapid development of industrial automation, higher requirements are put forward for reliable and deterministic communication in industrial networks. And time-sensitive networking (TSN) is a promising technology that can satisfy such deterministic transmission requirements. Currently, TSN typically uses the shortest path routing (SPR) algorithm to determine the transmission path of traffic. However, the SPR algorithm may cause a high load on a single path, which makes it difficult to improve the schedulability and determinism of time-triggered (TT) traffic. In this paper, a TSN traffic scheduling method based on Q-learning routing optimization for TT traffic is proposed, and the transmission performance of the proposed method is tested. The results show that the delay and jitter of TT traffic are reduced after using this method.