基于干扰观测器的欠驱动AUV自适应反演控制  

Adaptive backstepping control of underactuated AUV based on disturbance observer

作　　者：陈巍[1] 魏延辉[1] 曾建辉[1] 胡佳兴[1] 王泽鹏[1] 

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《中南大学学报（自然科学版）》2017年第1期69-76,共8页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(51205074);国家科学技术部国际科技合作项目(2014DFR10010);中央高校基本科研业务费专项资金资助项目(HEUCF041703);高等学校博士学科点专项科研基金资助项目(20112304120007)~~

摘　　要：针对未知外界干扰存在的情况,提出一种基于非线性干扰观测器(NDO)的自适应反演控制,用于改善欠驱动自治水下机器人(AUV)深度跟踪控制性能。首先对欠驱动自治水下机器人的垂直面模型进行有条件的简化,得到新的运动学及动力学方程,建立简化后模型的状态方程;其次根据系统状态方程构建非线性观测器,设计自适应反演控制器,输出控制量根据李雅普诺夫稳定原理推导得出,确保包含非线性干扰观测器及自适应反演控制器在内的控制系统的一致渐进稳定性。研究结果表明:所设计的非线性干扰观测器以及自适应反演控制器可以实现欠驱动自治水下机器人在存在外界未知干扰情况下深度轨迹跟踪,具有较强的鲁棒性。In order to improve the depth tracking control performance of underactuated autonomous underwater vehicle （AUV）, an adaptive backstepping controller based on nonlinear disturbance observer （NDO） was proposed considering unknown external disturbance. Firstly, the vertical plane motion model of underactuated AUV was simplified according to some limited conditions. The new kinematics and dynamics equations were obtained, and the state equations of simplified model were established as well. Secondly, on the basis of state equations, a nonlinear disturbance observer was constructed. In the meanwhile, an adaptive backstepping controller was designed based on NDO, and then the control output was obtained according to the Lyapunov stability principle, which globally guaranteed the asymptotical stability of the whole control system. The results show that designed NDO and adaptive backstepping controller can effectively achieve depth trajectory tracking of underactuated AUV in the case of unknown external disturbance, and show strong robustness to disturbance.

关 键 词：欠驱动自治水下机器人 深度跟踪 非线性干扰观测器 自适应 反演法 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]