【作者】 杨雪；

【导师】 唐功友；

【作者基本信息】 中国海洋大学 ， 计算机应用技术， 2014， 博士

【摘要】 21世纪以自主式智能体(Autonomous Intelligent Agents)为代表的机器人系统得到了人们的广泛关注。轮式移动机器人(Wheeled Mobile Robots, WMRs)和自主式水下航行器(Autonomous Underwater Vehicles, AUVs)是其中的典型代表。WMR和AUV分别能在陆地和水中自主执行各种复杂的任务，且在民用、工业及军事领域方面有广泛的应用。在执行海底石油管道和海底电缆维修检测、地形地貌探测、海洋观测以及军事应用等任务时都需要智能体能够高效精确地跟踪特定曲线，这使得研究如何实现自主式智能体的跟踪控制具有重要意义。其中AUV是一类比较复杂的非线性系统，状态向量之间相互耦合，控制难度较大。本文主要的研究内容和创新点现概括如下：1.研究WMR及全驱动AUV的点镇定问题。首先建立极坐标系下的系统方程，基于Lyapunov方法得到运动学控制器设计方法。其次，结合系统的动力学特性，利用Lyapunov方法和Backstepping法将运动学控制器拓展到动力学，设计出光滑的控制律。仿真实例验证设计方法的有效性。2.提出WMR在有限时间内的停车控制问题，即设计控制策略使机器人在有限时间内停靠在预先给定的停车区域内。基于极坐标系下机器人的非线性运动模型，将机器人运动控制系统分解为原地转动的线性系统及直线运动和转动结合的弱非线性系统两种形式。利用Lyapunov设计方法，得到了在有限时间内机器人到达给定停车区域的切换控制律，并给出了到达给定区域所需时间的估算公式。仿真算例说明控制方法的有效性。3.基于级联方法研究一类质心和几何中心不重合的WMR的自适应轨迹跟踪控制问题和海流影响下AUV的轨迹跟踪问题。将跟踪误差分解为位置和方位角误差两个子系统，利用Backstepping法和Lyapunov方法设计位置跟踪控制器，利用Lyapunov方法设计航向角跟踪控制器。对于WMR，针对参数未知的情况，通过自适应方法进行参数估计，给出自适应轨迹跟踪控制律的设计方法，并保证轨迹跟踪系统的稳定性。对于欠驱动AUV系统，给出海流影响下轨迹跟踪控制器的设计方法，同时给出了部分控制器参数的调整方法。4.研究海流影响及参数不确定条件下AUV的地形跟踪控制。基于Serret-Frenet坐标系，建立海流影响下的路径跟踪误差方程，并引入额外的系统自由度来达到控制目的。利用Lyapunov方法设计AUV的运动学跟踪控制器，然后采用Backstepping法得到动力学跟踪控制器，最后针对AUV参数未知的情况，给出其自适应控制方案。给出不同地形仿真实例与效果图，验证控制方法的可行性。5.研究时滞依赖的非线性AUV系统的H∞控制。通过对Fossen提出的AUV六自由度模型的简化，给出非线性四自由度的AUV数学模型，并考虑到时滞和海流、风浪等因素的影响，得到不确定时滞系统，采用线性矩阵不等式(LinearMatrix Inequality, LMI)方法和积分不等式技术得到时滞依赖的H∞控制律的设计方法，仿真结果表明所设计控制方法的有效性。更多还原

【Abstract】 21st century Autonomous Intelligent Agents have been widely concerned as therespresentative of the robot systems. Wheeled Mobile Robots (WMRs) andAutonomous Underwater Vehicles (AUVs) are typical ones of them. WMR and AUVcan autonomously perform complex tasks both on land and in water, respectively.They are widely used in civil, industrial and military fields. In the seabed oil pipelineand sunmarine cable detection, topograph detection, ocean observation and militaryapplications, intelligent agents should track a specific curve efficiently and accurately.It is of significant in reseaching the tracking control of autonomous intelligent agents.AUV is a class of complex nonlinear systems, mutual coupling between the statevectors. This increases the difficulty for the control of AUV. The major results andinnovations of this dissertation are summarized as follows.1. The point stabilizaion of WMR and full-actuated AUV are discussed. First, thesystem equations are established in polar coordinates. The designing methods ofkinematic controller are obtained. Next, the kinematic controller is extended todynamic controller through Lyapunov methods and Backstepping combined with thedynamic characteristics of the system. The smooth control laws are designed. Thesimulation results show the effectiveness of the control method.2. The parking problem of WMR in finite time is proposed. The control objectiveis to make the robot stop in a pre-given parking area. Based on the nonlinear motionmodel of the robot in polar coordinates, the motion control system of the robot isdivided into two types. The two types are the linear system of rotation and the weaklynonlinear system of rectilinear motion together with rotation. The switching controllaws are obtained through Lyapunov designing method. The control laws guaranteethe robot reach a given parking domain in finite time. The estimatin formula of thefinite time is obtained. A simulation example illustrates the effectiveness of the designing method.3. The problem of adaptive trajectory tracking control for WMR and the problemof trajectory tracking control for underactuated AUV based on cascaded method isresearched. There is a distance between the mass center and the geometrical center ofthe WMR. The tracking error system is divided into two subsystems. They are theposition and the orientation error system. Backstepping technique and Lyapunovmethod are applied to design the position tracking controller. Lyapounov method isapplied to design the orientation tracking controller. An adaptive tracking control lawis proposed to deal with the circumstance with the unknown parameters for WMR.The stability of the system of trajectory tracking is ensured. The designing method oftrajectory tracking controller for the underactuate AUV on the horizontal plane in thepresence of ocean currents is obtained. The adjustment methods of some controllerparameters are derived.4. The bottom-following control problem of underactuated AUV in the presenceof ocean currents and uncertain parameters is proposed. The Serret-Frenet frame isused to describe the bottom-following of the AUV. An extra degree of freedom forcontroller design is introduced. It make the virtual AUV can regulate its velocityalong with the real AUV. Based on Lyapunov method and Backstepping techniques,an adaptive controller is designed to ensure that the AUV converges to the desiredpath asymptotically. Different terrain simulations and effect diagrams are given toshow the effectiveness of the proposed controllers.5. Delay-dependentH∞control for nonlinear AUV is researched. Bysimplifying the Fossen’s six degrees model of AUV, the AUV’s four degrees model isobtained. The uncertain time-delay system is built through taking into accout theeffect of time delay and ocean currents, waves and other factors. Linear matrixinequality (LMI) and integral inequality techniques are applied to derivedelay-dependentH∞control laws. Simulation results illustrate the effectiveness ofthe control methods proposed.更多还原