【作者】 杜娟；

【导师】 郭书祥；

【作者基本信息】 哈尔滨工程大学 ， 控制理论与控制工程， 2013， 博士

【摘要】 随着水下机器人学及各种与机器人相关的科学技术的发展，水下机器人的研究已经取得了许多令人瞩目的成果，目前，世界上有许多国家正在致力于水下机器人的研究与开发，水下机器人的应用领域非常广泛，其应用领域涉及工业、渔业、探索和军事等等，水下机器人已经成为人们认识、开发和利用海洋的一个重要工具。水下机器人的控制是实现水下机器人运动的关键所在，水下机器人的控制问题涉及的方面非常多，水下机器人运动控制的难点主要在于它的工作环境为水下，而水下各种扰动和不确定因素的影响，以及实验环境的不便利势必会增添水下机器人的控制难度。使用一些传统的方法设计水下机器人控制器时，所获得的性能尚不能令人完全满意，因此有必要努力探讨高效的且易于实现的水下机器人运动控制方法。本论文将所研制的由喷水推进器和伺服电机组合的三套喷水推进系统作为矢量喷水推进方式的自主式水下球形机器人作为主要研究对象，进行相关的推进系统建模和动力学建模等研究，并在所建立模型的基础上研究出适合于该矢量喷水推进水下球形机器人的智能控制算法，该算法需要实现水下球形机器人在复杂海洋环境下的稳定运动控制，并且该算法需对模型要求不高、不过于复杂，以利于工程实现。本论文旨在选取对系统数学模型要求不高的模糊控制方法作为控制方法的主要研究对象，将自适应控制和滑模控制方法与其结合起来以设计出适合于水下球形机器人的控制器。论文的主要研究工作如下：第一，对所研发的自主式矢量喷水推进水下球形机器人进行了机械设计，然后对水下球形机器人内部控制系统进行了电气设计，使人们从宏观上了解自主式矢量喷水推进水下球形机器人的整体架构和工作原理。第二，详细地讨论了水下球形机器人推进系统的设计与分配以及合推力计算，并分别对单个喷水推进系统和多个喷水推进系统进行了建模与辨识研究以实现推进系统的推力预报，推力预报则主要采用实验测量和理论分析相结合的方法，该部分所得到的喷水推进系统模型为水下球形机器人动力学模型中的控制输入矢量。第三，在流体力学基本概念的基础上对水下机器人的运动学和动力学进行了分析，得到了水下机器人总的动力学模型，根据水下机器人运动学和动力学分析结果，针对矢量喷水推进水下球形机器人这一特殊形体进行了动力学声明，进而实现了必要的简化计算，最终得到了该水下球形机器人的简化动力学模型，并对该模型进行了相关的仿真实验，将仿真结果与实际水下实验结果对比以验证所建立模型的有效性；对水下球形机器人前进、升沉和转艏这三个自由度的运动进行了相关的流体力学特性分析，得到了水下球形机器人球体内结构对速度和压力的影响情况。第四，在所建立的喷水推进器和水下球形机器人的模型的基础上，对水下球形机器人进行了相关的控制算法研究，力求探究得到合适的运动控制方法，先对水下球形机器人水平方向上的前进运动进行了PID控制和模糊滑模控制的研究，再基于模糊控制理论、自适应控制理论和滑模控制理论，将其有机结合，针对水下球形机器人提出了改进的自适应模糊滑模控制器，并进行了有扰动存在下的仿真研究，同时，将其结果与所设计的PID控制器和模糊滑模控制器的仿真结果进行了对比，以证明所设计的改进的自适应模糊滑模控制器是否优越。第五，采用基于传感器的反馈信息在模糊控制思想的基础上可在线调整的控制器，对水下球形机器人经常用到的前进、升沉和转艏这三个自由度的运动进行了一系列的水下运动实验，并对多个机器人的共同前进运动进行了水下运动实验，以验证该水下球形机器人的可行性和有效性。更多还原

【Abstract】 With the development of underwater robotics and robot-related science and technology,the researches on underwater robots have made many remarkable achievements, at present,there are many countries in the world are committed to the research and development ofunderwater robots. Underwater robots show significant potential of being applied in the fieldsof industry, fishery, exploration, and military and so on. Underwater robots have become oneof the most important tools to exploit and use marine resources. The control of underwaterrobots is one of the key technologies for these kinds of robots, and the control problems ofunderwater robots involve many aspects. The difficulty of underwater robot control lies in itsworking environment is underwater, underwater various disturbances and uncertainties, aswell as the inconvenience of the experimental environment is bound to add the difficulty ofunderwater robot control. The resultant performance cannot satisfy our requirements whenusing some traditional methods to design the controller of underwater robots. Therefore, it isnecessary to investigate more efficient and easy to implement motion control methods.The developed autonomous spherical underwater robots will be seemed as an object ofstudy, and the three sets of water jet propulsion system which is the combination of the waterjet thrusters and servo motors were as the way of vectored water jet propulsion. The relatedstudy about propulsion system modeling and dynamic modeling will be carried on, and got anintelligent control algorithm which is fit for the vectored water jet propulsion sphericalunderwater robots on the basis of the model. The control algorithm should be more suitable toimplement a stable motion control in the complicated underwater environments; also, itshould be simple and has a low requirement of the model, which can be easily implemented inproject applications. This paper aimed to choose the fuzzy control as the main research point,and combined the self-adaptive control and sliding mode control to design a controller whichis more suitable for the spherical underwater robots.The main research work of this paper are as follows: First, we carried mechanical designon the developed autonomous vectored water jet propulsion spherical underwater robots, andwe also carried electrical design on the inner control circuit of the spherical underwater robots,so that people can understand the overall framework and working principle of the autonomousvectored water jet propulsion spherical underwater robots. Second, we discussed the designand distribution of the propulsion system in detail, and we calculated the resultant actuatingforce, also we carried on modeling and identification research on a single and multiple water jet propulsion systems respectively in order to realize the thrust forecast of the propulsionsystems. The thrust forecast mainly uses the method of combining experimentalmeasurements and theoretical analysis, and the derived water jet propulsion system model isthe control input vector of underwater spherical robots dynamic model. Third, we analyzedthe kinematics and dynamics of the underwater robots on the basic concepts of the fluidmechanics, and built the dynamics model of the whole system. According to the analysisresults of kinematics and dynamics, we carried the dynamics statement on the vectored waterjet propulsion spherical underwater robots, thus we realized the necessary calculationssimplified and gained the simplified dynamics model based on the spherical underwaterrobots finally. We also carried on model simulation, and compared the simulation results withthe actual underwater experimental results in order to verify the validity of the model. Inaddition, we discussed the hydraulics analysis of three underwater motions, surge, heave, andyaw of the spherical underwater robots and obtained the impact of the spherical underwaterrobots structure on velocity and pressure. Fourth, we studied the control algorithm of thespherical underwater robots based on the model of water jet thrusters and sphericalunderwater robots, and we carried research on the PID control and the fuzzy sliding modecontrol for forward movement of spherical underwater robots firstly, then based on the fuzzycontrol, the self-adaptive control, and the sliding mode control, we proposed an improvedadaptive fuzzy sliding mode controller for the developed spherical underwater robots, and wecarried on simulation studies with disturbance, at the same time we compared its results withthe simulation results of the designed PID controller and fuzzy sliding mode controller inorder to prove the excellence of the improved adaptive fuzzy sliding mode controller. Finally,we carried out a series of experiments on three degrees of freedom of movement, surge, heave,and yaw, by adopting sensor-based feedback controller which can be adjusted online based onfuzzy control idea. Also, we carried on the common surge motion experiments of multipleunderwater robots in order to prove the feasibility and validity of the developed sphericalunderwater robots.更多还原