Solving a Class of Mean-Field LQG Problems

• 成果类型: Article
• 署名作者: Li, Yun; Song, Qingshuo; Wu, Fuke; Yin, George
• 署名单位: Huazhong University of Science & Technology; Worcester Polytechnic Institute; University of Connecticut
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2021.3073204
• 发表日期: 2022
• 页码: 1597-1602
• 关键词: Cost function; optimal control; Extraterrestrial measurements; Aerospace electronics; trajectory; STANDARDS; PROCESS CONTROL; Controlled diffusion; linear quadratic Gaussian (LQG) control; McKean-Vlasov equation; partially observable system

摘要

In this article, we study a class of mean-field linear quadratic Gaussian problems. Under suitable conditions, explicit solutions of the distribution-dependent optimal control problems are obtained. Riccati systems are derived by directly solving the associated master equations. Some extensions on controls with partial observations are also considered.