Nonlinear dynamics of novel flight-style autonomous underwater vehicle with bow wings, Part I: ASE and CFD based estimations of hydrodynamic coefficients, Part II: Nonlinear dynamic modeling and experimental validations
Faheem Ahmed,Xianbo Xiang,Haotian Wang,Jialei Zhang,Gong Xiang,Shaolong Yang
DOI: https://doi.org/10.1016/j.apor.2023.103739
IF: 3.761
2023-01-01
Applied Ocean Research
Abstract:Accurate dynamic modeling is essential for precise prediction of six degrees-of-freedom (6-DOF) motion and model-based controller design of an autonomous underwater vehicle (AUV). The accuracy of dynamic model of an AUV largely depends upon the precise estimation of hydrodynamic coefficients. Accordingly, to establish the dynamic model based on ASE-estimated hydrodynamic coefficients of the novel flight-style AUV and experimental validations, the study is presented in two parts (Part I: ASE and CFD based estimations of hydrodynamic coefficients and Part II: Nonlinear dynamic modeling and experimental validations). Part I: In this part complete set of nonlinear hydrodynamic coefficients including, added mass, drag, and lift forces and moment coefficients are estimated using ASE technique, which is specifically developed to accurately comprehend the novel AUV design including, AUV hull, fixed wings, control fins, and appendages. Additionally, an improved ASE approach is also proposed to estimate cross flow drag force coefficients. In proposed improved ASE technique, (C-dc) in mathematical formulation is taken from the experimental results of 3D flow past over circular cylinder instead of 2D, as followed previously to estimate the nonlinear cross-flow drag force and moment coefficients (Y-v|v|,Z(w|w|),M-w|w|,N-v|v|,Y-r|r|,Z(q|q|),M-q|q|&N-r|r|). The proposed ASE technique is verified and validated in two stages: first, by comparing the ASE-estimated results with CFD-based estimations, and second, by investigating the 6-DOF motion response of the AUV. When compared ASE-estimations with the CFD results for the cross-flow drag force and moment coefficients, the proposed ASE technique has shown a significant improvement (approximately 20%) over the previously employed ASE method. The ASE-estimated results are validated through 6-DOF motion response of AUV during free-running underwater experiments, provided in 'Part II'. Overall, the proposed ASE technique is considered more effective, direct, and less costly in terms of time and computational efforts, compared to CFD-based or experimental-based estimations of hydrodynamic coefficients of AUVs, particularly torpedo-shaped AUVs. Part II: This part presents a nonlinear dynamic model for a newly designed torpedo-shaped flight-style AUV having complex bow and stern fixed wings and control fins configuration. The model includes precise mathematical formulations to accurately represent the physics of the AUV and underwater environment in terms of hydrodynamic derivatives such as added mass, drag, and lift forces and moment coefficients. These coefficients have been estimated using an ASE method provided in 'Part I'. The 6-DOF motion response of the AUV has been simulated in both horizontal and vertical planes which includes, zigzag yaw, circle, zigzag pitch, and spiral maneuvers. The lift generated by the bow and stern control fins has been investigated both separately and collectively, especially during complex spiral ascending maneuvers. Moreover, improved ASE technique for cross flow drag coefficients has also been validated through 6-DOF free-running underwater experiments of AUV and insights on impact of cross flow drag coefficients on 6-DOF motion response of AUV has also been provided. Overall, the comparison between the simulated AUV motion and free-running underwater experiments demonstrated consistency, thereby validating the accuracy of the nonlinear dynamic modeling, ASE-based hydrodynamic coefficient predictions, and effectiveness of the newly designed flight-style AUV for underwater operations.