Adaptive multi-body yawing filter of wave driven robot with dynamic response amending. (15th December 2021)

• Record Type: Journal Article
• Authors: Liao, Yulei; Li, Ye; Pan, Kaiwen; Wang, Leifeng; Zhang, Weixin
• Is Part Of: Ocean engineering. Volume 242(2021)
• Publication Date: 2021-12-15
• Subjects: Adaptive filter -- Kalman filter -- Linearized -- Multi body -- Response model -- Wave driven robot; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2021.110167
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: Wave driven robots (WDRs) are a new type of autonomous ocean observation platform driven by wave energy. This paper is concerned with the multi-body yawing filter problem of the WDR. The adaptive multi-body yawing filter (AMF) of WDR was presented by considering the driven principle of the yawing motions of the submerged glider (glider) and the float body (float). The linearized yawing response models of the glider and the float were derived. The adaptive amending method of the dynamic response coefficients (DRCs) of the glider and the float were proposed. On this basis, the AMF of WDR was developed within the extended Kalman filter (EKF) framework. The AMF method utilized input and output (I/O) data to extract model information and is independent of the accurate mechanism model of the system. Numerical simulations of step steering and continuous steering under nominal model and the model with parameter perturbation were performed. The sea trial was conducted and the filtering results of the actual sensor data by AMF, EKF and tracking differentiator (TD) were presented. The proposed AMF method was found to be with good performance in suppressing noise and tracking speed. Highlights: The adaptive multi-body yawing filter (AMF) of wave driven robot (WDR) was presented by considering the driven principle of the yawing motions of the glider and the float. The adaptive amending method of the dynamic response coefficients (DRCs) of the glider and the float were proposed which is independent of the mathematical model. Numerical simulations showed that the proposed AMF suppresses the heading sensor noise well under both nominal model and model with parameter perturbation. Sea trials showed that the filtering results of AMF were smooth, unbiased and almost without lag.