Scholarlink

<p>The control of underwater vehicle-dual-manipulator (UVDM) systems is constrained by severe nonlinearities, coupling effects, and external disturbances. Existing robust control strategies often suffer from key limitations, including the use of conservative, constant upper bounds for disturbances, the high-frequency chattering associated with standard sliding mode control, the neglect of actuator saturation, and the model-dependence of conventional disturbance observers (e.g., reliance on the inertial matrix of system). To address these challenges, this paper presents two enhanced robust adaptive control frameworks. The first strategy is an enhanced adaptive sliding mode control (ASMC), which serves as a model-based benchmark. This method employs a rigorous adaptive law to estimate the upper bounds of disturbances, first considering them as an unknown constant and then, more realistically, as a nonlinear function of the system states. Crucially, the ASMC replaces the discontinuous sign function with a continuous adaptive term, thereby eliminating chattering while maintaining strong robustness. The second strategy is a robust adaptive model-free disturbance observer control (MFDOC). Its principal merit lies in its integrated disturbance estimation mechanism, which operates without an explicit model of the plant dynamics. Furthermore, this controller is explicitly augmented to handle input saturation. By eschewing the need for a separate, model-dependent observer design phase and mitigating chattering through the observer approach, the MFDOC offers a structurally simpler solution robust against plant uncertainties. A comprehensive comparative study evaluates the performance and applicability of both controllers. The ASMC provides precise tracking and stability, showcasing the power of adaptive techniques within a traditional control structure. The MFDOC excels in structural simplicity and independence from complex system models, making it highly practical for systems where accurate modeling is challenging. The stability of each scheme is rigorously proven using Lyapunov analysis. Simulation results show that the proposed ASMC reduces the integral of absolute error (IAE) by approximately 79% and 70%, the integral of time-weighted squared error (ITSE) by about 91% and 83%, and the control energy (CE) by approximately 85% and 94%, for the constant-bound and functional-bound cases, respectively, compared with a conventional sliding-mode controller, demonstrating their superior tracking accuracy, robustness, and control efficiency.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Two Enhanced Robust Adaptive Control Strategies for Uncertain Underwater Vehicle-dual-manipulator Systems

  • 期刊论文

  • 发表日期：

    2026年3月26日

查看全文
  • Seyed Omid Afzali,
  • Alireza Alfi,
  • Javad Keighobadi

摘要

The control of underwater vehicle-dual-manipulator (UVDM) systems is constrained by severe nonlinearities, coupling effects, and external disturbances. Existing robust control strategies often suffer from key limitations, including the use of conservative, constant upper bounds for disturbances, the high-frequency chattering associated with standard sliding mode control, the neglect of actuator saturation, and the model-dependence of conventional disturbance observers (e.g., reliance on the inertial matrix of system). To address these challenges, this paper presents two enhanced robust adaptive control frameworks. The first strategy is an enhanced adaptive sliding mode control (ASMC), which serves as a model-based benchmark. This method employs a rigorous adaptive law to estimate the upper bounds of disturbances, first considering them as an unknown constant and then, more realistically, as a nonlinear function of the system states. Crucially, the ASMC replaces the discontinuous sign function with a continuous adaptive term, thereby eliminating chattering while maintaining strong robustness. The second strategy is a robust adaptive model-free disturbance observer control (MFDOC). Its principal merit lies in its integrated disturbance estimation mechanism, which operates without an explicit model of the plant dynamics. Furthermore, this controller is explicitly augmented to handle input saturation. By eschewing the need for a separate, model-dependent observer design phase and mitigating chattering through the observer approach, the MFDOC offers a structurally simpler solution robust against plant uncertainties. A comprehensive comparative study evaluates the performance and applicability of both controllers. The ASMC provides precise tracking and stability, showcasing the power of adaptive techniques within a traditional control structure. The MFDOC excels in structural simplicity and independence from complex system models, making it highly practical for systems where accurate modeling is challenging. The stability of each scheme is rigorously proven using Lyapunov analysis. Simulation results show that the proposed ASMC reduces the integral of absolute error (IAE) by approximately 79% and 70%, the integral of time-weighted squared error (ITSE) by about 91% and 83%, and the control energy (CE) by approximately 85% and 94%, for the constant-bound and functional-bound cases, respectively, compared with a conventional sliding-mode controller, demonstrating their superior tracking accuracy, robustness, and control efficiency.