节点文献

介入诊疗机器人多体系统建模分析与实现

Multi-Body Systems Modeling, Analysis and Implementation of Interventional Diagnostic Micro-Robots

【作者】 赵大旭

【导师】 吴洪涛;

【作者基本信息】 南京航空航天大学 , 机械电子工程, 2010, 博士

【摘要】 随着生物工程、微电子机械技术(MEMS)的发展和微系统加工技术的成熟,用于介入诊疗手术的介入诊疗机器人,已经成为国内外研究的热点问题。人体管腔是个复杂的管道系统,不仅几何形状复杂,而且内部存在生物液流场。机器人在介入诊疗过程中的受力情况相当复杂。介入诊疗机器人在人体管腔这样的特殊环境中作业,必须保证绝对安全和可靠。在此前提下,机器人要在有限尺寸限制下集成尽可能多的功能模块,并保证运动灵活性,基于力学模型的设计与控制尤为重要。本论文建立了主动介入机器人典型运行环境——动脉血管和消化道的数学模型,并进行了数值模拟,在此基础上基于多体系统动力学理论,研究了介入诊疗机器人在管道几何形状约束和流场作用下的运动学和动力学问题,论文的研究结果对包括水下机器人在内的液体环境下作业的管道机器人研究也具有重要的参考价值。由于没有固定基座,介入诊疗机器人的运动与人体管腔以及管内流场运动相互耦合,系统整体受人体管腔几何形状约束,姿态变化范围较大,因此其动力学问题相对复杂,在分析了典型人体管腔的基础上,引入Euler四元数表达机器人的姿态,解决了模型数值计算中的奇异性问题。机器人所有的位置姿态构成一个Lie群,利用Lie群和Lie代数描述系统是很自然和方便的,Lie代数的本质是运动旋量。旋量理论以及空间算子代数理论的引入,使机器人运动参数的计算具有统一的简洁形式。论文将游动型介入诊疗机器人简化为管道流场环境下单个典型体的六维空间运动问题,建立了游动机器人在血流冲击下的运动学及动力学模型,对游动机器人在三维弯曲管道内沿规划路径的游动特性进行了仿真分析,为机器人结构设计和运动控制提供了依据。论文通过CFD软件计算出具体环境下的流场力,并将受力以载荷形式代入动力学模型,简化了系统的动力学模型而又不失问题的本质,在单个典型体动力学基础上,将问题拓展到更复杂拓扑结构的系统中。针对更一般的介入诊疗机器人系统,假设系统不但包含刚体,也包含柔性体,而且既可以是链式系统又可以是树型多体系统,本论文分析了开链式拓扑结构的多节蠕动机器人结构和运动特点,利用Huston理论的低序体阵列方法描述系统的拓扑结构,然后根据判断相邻两体是刚体或者柔体,利用空间算子代数(SOA)方法,建立了适用于链式和树形蠕动机器人系统的通用动力学模型,并进行了初步验证试验。论文进一步发展了空间算子代数理论体系,采用空间算子处理了介入诊疗机器人多体系统动力学高效率建模问题。根据系统中铰的驱动情况分别对铰链定义为主动铰和被动铰,通过判断铰链的类型分别按照两次从系统的顶端到根体的顺序、一次从根体到末端的顺序进行了系统铰接体惯量的递推、系统冗余力的递推和广义加速度和广义主动力的递推。通过上述递推过程建立了主动介入机器人系统广义递推动力学模型,实现了高效率O(n)次的计算效率,该算法可以应用于其他管道机器人或者水下机器人(包括刚性多体系统、柔性多体系统、欠驱动系统),求解反向动力学、正向动力学和混合动力学递推。利用所建立动力学模型,论文分析了多节蠕动型介入诊疗机器人。课题组根据仿生学原理,分别研制了主动型的游动和蠕动介入诊疗机器人,在专门构建的模拟测试平台上,对机器人驱动机理进行了试验研究,对动力学参数进行了识别,得到驱动力与流场阻力等数据。在此基础上,运用Mathematica符号软件开发了专门的仿真程序,对典型的介入诊疗任务进行了仿真研究,得到机器人相关的驱动力信息,以及关节运动信息,对本文提出的理论方法进行了验证。

【Abstract】 Vascular interventional robot has been become a hot scientific research topic currently at home and abroad with the development of Biotechnology Engineering, MEMS and its processing technology. Human lumen is a complex pipeline system, not only in geometry but also there are biological flow field inside it. The force situation of the vascular interventional robot is very complex.Because vascular interventional robot will run in humen lumen that so special envorement that it must be absolute security and reliability. Under the presupposition, to integrate functional modules in the limited of size constraints as much as possible and ensure flexibility of movement, it is particularly important for designing and controlling on the base of the mechanical model. The typical operating environment, the mathematical model of the arteries and digestive tract, is built where the robot will run, and is simulated as a numerical model. Then based on this multi-body dynamics, the kinematics and dynamics theory of the interventional robot are studyed based under constraints of pipe geometry and the envorement of flow field. The paper’s findings have great reference value in researching on underwater robots and pipe robot which will be operated in liquid field.There are many factors caused complex in analysing the dynamic problems, such as no fixed base, coupling motion of intervention robot with the motion of the liquid in humen lumen, the size limited of humen lumen, the more varies of pose, e.t. Expressing robot pose is cited the euler quaternions which solve the singularity of the numerical model well. All positions and orientations of the robot form a Lie group. Using Lie group and Lie algebra to describe multi-mechanical system is very convenient. Lie algebra is the essence of motion screw. Introduction of screw theory and spatial operator algebra theory lead that it a unified simple form in the calculation of the motion parameters. Intervention robot is simplified as a single typical six-dimensional space mobility in this paper, and it’s kinematics and dynamics model of running in the blood flow. Then robot is simulated in the three-dimensional dust running along the palnning path, that provides a basis for designing structure and controlling motion. The liquid force on robot in the specific environment is calculated by CFD software and the force is substituted in the dynamic model as the form of load. This simplifies the dynamic model and not losing the essence of the problem. This method is extended to use in the more complex topology system on the basis of single typical body dynamics.For more general interventional robotic systems, assuming that the system contains not only the body, also contains flexible body, and maybe either chained system and tree-type multi-body systems. This paper analyzes the structure and motion characteristics of the open-chain topology multi-section creeping robot, and uses the theory Huston lower body array method to describe the system topology. Then according to the judgment on differentiating rigid or flexible body, using space operator algebraic (SOA) method, establish a commen dynamics model of the chain and common tree creeping robot, and carry on some experiments to verify the results. The paper further develops the theoretical system of spatial operator algebras and describes the high-efficiency modeling of the general dynamics of active interventional multi-body system by use of spatial operator. Hinges are defined as active hinge and passive hinge respectively, according to driving types of hinge in the system. Responding to the types of hinge, the recursion of hinge’s inertia, general acceleration and general active force, and redundant force of the system are performed in the order that from the top of the system to the base and from the base to the top respectively. Through the three methods of recursion mentioned above, the paper has set up the general recursion dynamics modeling of the active interventional robot system and achieved high computation efficiency of O(n). This arithmetic can be applied to the recursion of backward, forward and compounded dynamics of pipe and underwater robots (excluding rigid and flexible multi-body system and drive-lack system). Using the dynamic model established before, the paper analyzes the multi-section creeping interventional robot.Research group have manufactured the squirm and nomadic robots, according to the bionic principle. Studying the drive mechanism of the robot, realized the recognition of the robot’s dynamic parameters on the simulated platform that has been established, and finally get the data value of the driving force and the resistance of the flow field, etc. On this basis, the paper has developed a special simulation program with symbol software of Mathematica and conducted a simulation research on the typical interventional assignment. And finally, the paper has got the driving force information related to the robots and the motion information of the joints and verified the theory and methods claimed in the paper.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络