Deep Reinforcement Learning for Continuous-time Self-triggered Control⁎This work was supported in part by JSPS KAKENHI under Grant Number JP18H01461 and JP21H04875. Issue 14 (2021)

• Record Type: Journal Article
• Authors: Wang, Ran; Takeuchi, Ibuki; Kashima, Kenji
• Is Part Of: IFAC-PapersOnLine. Volume 54:Issue 14(2021)
• Page Start: 203
• Page End: 208
• Publication Date: 2021
• Subjects: Self-triggered control -- data-driven control system design -- machine learning; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2021.10.353
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: In recent years, the trade-off between communication cost and control performance has become increasingly important. Among various control architectures, self-triggered controllers decide the next communication (state observation and action determination) timing online in a state-dependent manner. However, it should be emphasized that most of the existing methods do not explicitly evaluate the resulting long-run communication cost. In this paper, we formulate an optimal continuous-time self-triggered control problem that takes the communication cost into an explicit account and proposes a design method based on deep reinforcement learning.