Implementation and Evaluation of Non-linear Optimal Feedback Control for Ship's Automatic Berthing by Recurrent Neural Network. Issue 21 (2019)

• Record Type: Journal Article
• Authors: Mizuno, Naoki; Kuboshima, Ryo
• Is Part Of: IFAC-PapersOnLine. Volume 52:Issue 21(2019)
• Page Start: 91
• Page End: 96
• Publication Date: 2019
• Subjects: ship control -- automatic berthing -- non-linear optimal control -- feedback control -- recurrent neural network -- machine learning; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2019.12.289
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: In this paper, we present an automatic ship's berthing system by non-linear optimal feedback controller. In the proposed method, the recurrent neural network is used for non-linear optimal feedback controller for the realistic operational conditions such as different berthing distances and different disturbances. To obtain the feasible non-linear controller, the recurrent neural network is trained using pre-computed non-linear optimal solutions for various conditions. In order to evaluate the performance of the proposed system, extensive computer simulations and actual sea tests are carried out using small training ship Shioji-Maru under various conditions. As a result, we can see that the proposed non-linear optimal feedback controller by recurrent neural network is useful for automatic berthing system for the ship.