类Lyapunov理论在类人形机器人任务空间内跟踪的应用

考虑了类人形机器人的各种不确定因素,提出了其手臂控制的新方法.基于类Lyapunov方法,设计出具有鲁棒性功能的任务空间控制器.该方法得到的控制器不但在有限确定时间内达到稳定跟踪,而且不需要雅可比矩阵求逆,对一定类型的外部干扰具有鲁棒性功能.最后通过数值仿真显示了所得结果的有效性及其应用方法.

拟人机器人交互作用力的研究

拟人机器人的动力学具有高度非线性、高度耦合的特点,分析清楚各组成部分之间的交互作用力是实施高级控制方法的基础。文章在以往分析移动机械手的基础上,从整体建模的角度入手,对拟人机器人的交互作用力提出了一个新的模型,即神经网络模型。利用该模型对一个特殊的单一手臂运动的例子进行了拟合,其结果是收敛的。这说明提出的模型是有效的,此后,我们将陆续给出研究成果。

空间柔性双臂机器人系统建模、控制与仿真研究

空间柔性多臂机器人系统是高度非线性,强耦合的动力学系统,其动力学的研究是比较复杂和困难的问题.本文针对极为复杂的空间柔性双臂机器人系统,建立了其系统动力学模型,利用逆动力学控制算法对该动力学模型实现了轨迹跟踪控制,仿真结果表明该方法具有较好的控制效果,分析了在仿真过程中出现的有关数值算法的问题.

高速重载工业机器人臂机构研究

研究出一种新的机器人臂机构———“并联放大”复合臂机构 ,介绍了其机构原理、结构设计及测试实验。实验表明 :载荷 2 5kg时 ,臂机构末端运动速度可达到 10m/s ,该臂机构具有高速运动及抗冲击振动的能力。

装配机器人系统快速碰撞检测算法

提出一种面向操作手装配系统的快速碰撞检测算法。该算法以机器人运动学和空间解析几何为基础 ,将判断机械手手臂与障碍物是否发生碰撞问题转化为直线段与有界平面是否存在公共点的简单解析几何问题 ,并以 PU MA5 6 0操作手为例对算法加以说明。该算法不仅适用于静态的障碍物已知的环境 ,而且适用于障碍物运动规律已知的动态环境 ,减少了碰撞检测占用的时间 ,提高了路径规划的效率

微型ROV控制装置及控制方法

介绍了一种基于嵌入式ARM9技术的微型ROV的控制装置及控制方法。该装置可以同时进行两通道串行通讯,实现微型ROV的视频信号、潜水深度、艏向角度、纵倾角度、横摇角度、电子舱温度等数据的采集和与上位机的通讯传输;该装置可以采集16路模拟量信号和12路数字量信号,输出4路模拟量信号和12路TTL电平信号,实现推进器、水下灯、水下摄像机、云台等ROV功能器件的驱动。该装置具有通讯能力强、集成度高、功耗低等特点,可以满足微型ROV所有的常用功能要求。

轮桨腿一体化两栖机器人控制系统研究与设计

轮桨腿一体化两栖机器人是一种既可以在陆地、滩涂、海底爬行，又可以在极浅水海域浮游的特种机器人。其研究与开发，对海洋环境调查、滩涂沼泽研究取样、军事两栖探雷与灭雷以及水下传感网络等都具有重要意义。这种新型两栖机器人作业环境复杂、运动模式多样，其运行机理、控制策略和建模方法等关键技术与传统的陆地或水下机器人有所不同。轮桨腿一体化两栖机器人控制系统的研究与设计，对轮桨腿一体化两栖机器人样机的实现及各项关键技术的进一步研究具有重要意义。 本文以轮桨腿一体化两栖机器人样机功能实现为目的，针对机器人载体具有的模块化、可重构和可扩展等特性，采用开放式设计思想，对控制系统软、硬件进行开放式设计。硬件上，采用基于CAN总线的分布式控制系统，便于实现功能节点的扩展，并且采用片内集成CAN控制器的ARM处理器，便于实现与CAN总线接口且具有较低的功耗。软件上，控制软件建立在μC/OS-II嵌入式实时操作系统之上，采用分层和模块化结构，以实现软件系统的开放性。同时，针对机器人载体空间狭小、难以布置下外购电机驱动器的特点，自行设计了体积较小且驱动能力较强的专用电机驱动电路。 第1章简要介绍研究的目的和意义后，分别对两栖机器人和机器人控制系统的国内外研究现状进行了综述。第2章介绍了轮桨腿一体化两栖机器人的系统组成，包括外形结构、硬件系统结构和软件系统结构，并结合两栖机器人自身特点，介绍了其运动控制方式。第3章对轮桨腿一体化两栖机器人的硬件系统进行了研究和设计，包括节点处理器选择、主控制节点设计、CAN数据总线接口电路设计、电机驱动节点设计、故障诊断及传感器节点设计、系统电源管理方案设计六个部分。第4章针对轮桨腿一体化两栖机器人的特殊需求，制定了CAN总线应用层协议，并采用嵌入式实时操作系统进行了通信软件和电机驱动软件的设计。第5章设计了控制系统在系统编程(ISP)方案。第6章对所设计的控制系统进行了实验验证。 实验平台的测试表明，本文设计的控制系统能够可靠、稳定工作；控制系统和控制方法的设计与实现是正确并且有效的。控制系统软硬件的设计对轮桨腿一体化两栖机器人样机的实现和各项关键技术的研究奠定了良好的基础。

两栖机器人控制系统研究与实现

轮桨腿一体化两栖机器人是一种既可以在陆地、滩涂、海底爬行，又可以在 极浅水海域浮游的特种机器人。它巧妙地将水下机器人常用的螺旋桨推进器与陆 地爬行机器人的驱动轮结合为一体、将水下方向舵与陆地爬行腿结合为一体，实 现在水下浮游与陆地爬行运动模式之间自动的切换，从而能够在普通陆地爬行机 器人和水下浮游机器人无法进入的极浅水、碎浪带和海滩区域进行考察作业。 这种新型两栖机器人作业环境复杂、运动模式多样，其运行机理、控制策略 和建模方法等关键技术与传统的陆地或水下机器人有所不同。本文对轮桨腿一体 化两栖机器人控制系统进行研究，针对机器人具有的模块化、可重构和可扩展等 特性，采用开放式设计思想，完成了具有混合体系结构特点的控制系统设计，并 从硬件和软件两个方面对控制系统的实现进行了阐述。 在体系结构设计上，整个控制系统分为三层：慎思层、序列层和行为层。其 中慎思层负责基于推理的计算与决策，实现两栖机器人高层智能性；行为层负责 两栖机器人基于反应的反馈控制，其包容结构实现了两栖机器人对动态环境的快 速响应能力和适应能力；连接慎思层和行为层的序列层根据两栖机器人自身状态， 将慎思层的决策结果转化为行为层的执行序列。 在硬件设计上，控制系统由水面控制台子系统和载体控制子系统组成。水面 控制台子系统接收人的使命指令，分解成任务序列，通过CAN 总线或无线电下载 到载体控制子系统，实现慎思层的功能。载体控制子系统是由嵌入式主控计算机 节点、6 个驱动装置控制器节点、故障诊断与应急处理节点构成的基于高速CAN 总线的分布式控制系统，它根据接受到的任务目标，以及环境状态和机器人内部 状态，完成对机器人的实时控制，实现了整个系统序列层和行为层的功能。 在软件设计上，使用ANSI C 语言完成了各种硬件驱动设计、部分功能模块设 计工作。分布式控制系统各嵌入式节点在μC/OS-II 嵌入式实时操作系统的基础上 完成软件设计，并实现与其他节点之间的CAN 通信。 实验表明，本文设计的控制系统能够可靠、稳定工作，并具有良好的模块化 结构和开放性，为轮桨腿一体化两栖机器人样机的实现和各项关键技术的研究奠 定了良好的基础。

双机械臂协调系统的一类最优轨迹规划方法

黄东海及朝鲜海峡冰后期陆架沉积物的环境磁学研究

There are four chapters in this dissertation. The first chapter briefly synthesizes the basic theories, methods and present-day applying situation of environmental magnetism. The second chapter probes into the magnetic mineral diagenesis in the post-glacial muddy sediments from the southeastern South Yellow Sea and its response to marine environmental changes, using the muddy sediment of Core YSDP103 formed in the shelf since about 13 ka BP. The third chapter illustrates the high-resolution early diagenetic processes by investigating the rapidly deposited muddy sediments during the last 6 ka in Cores SSDP-102 and SSDP-103 from the near-shore shelf of Korea Strait. The fourth chapter presents the results of detailed rock magnetic investigation of the surface sediments from the fine-grained depositional area on the outer shelf of the East China Sea in an attempt to provide environmental magnetic evidence for the provenance of the fine-grained deposit. Core YSDP103 was retrieved in the muddy deposit under the cold eddy of the southeastern South Yellow Sea, and the uppermost 29.79 m core represents the muddy sediments formed in the shelf since about 13 ka BP. The lower part from 29.79 to 13.35 m, called Unit A2, was deposited during the period from the post-glacial transgression to the middle Holocene (at about 6 ~(14)C ka BP) when the rising sea level reached its maximum, while the upper part above 13.35 m (called Unit Al) was deposited in a cold eddy associated with the formation of the Yellow Sea Warm Current just after the peak of post-glacial sea level rise. For the the uppermost 29.79 m core, detailed investigation of rock-magnetic properties and analyses of grain sizes and geochemistry were made. The experimental results indicate that the magnetic mineralogy of the core is dominated by magnetite, maghemite and hematite and that, except for the uppermost 2.35 m, the magnetic minerals were subject to reductive diagenesis leading to significant decline of magnetic mineral content and the proportion of low-coercivity component. More importantly, ferrimagnetic iron sulphide (greigite) is found in Unit A2 but absent in Unit Al, suggesting the control of marine environmental conditions on the magnetic mineral diagenesis. Magnetic parameters show abrupt changes across the boundary between the Unit Al and A2, which reflects a co-effect of environmental conditions and primary magnetic components of the sediments on the diagenesis. Alternating zones of high and low magnetic parameters are observed in Unit A2 of Core YSDP103, which is presumably due to periodic changes of the concentration and/or grain size of magnetic minerals carried into the study area. Cores SSDP-102 and SSDP-103, two studied sediment cores from the Korea Strait contain mud sequences (14 m and 32.62 m in thickness) that were deposited during the last 6,000 years. Analyses of grain sizes and geochemistry of the cores have demonstrated that the sediments have uniform lithology and geochemical properties, however, marked down-core changes in magnetic properties suggest that diagenesis has significantly impacted the magnetic properties. An expanded view of early diagenetic reactions that affect magnetic mineral assemblages is evident in these rapidly deposited continental shelf sediments compared to deep-sea sediments. The studied sediments can be divided into four descending intervals, based on magnetic property variations. Interval 1 is least affected by diagenesis and has the highest concentrations of detrital magnetite and hematite, and the lowest solid-phase sulfur contents. Interval 2 is characterized by the presence of paramagnetic pyrite and sharply decreasing magnetite and hematite concentrations, which suggest active reductive dissolution of detrital magnetic minerals, the absolute minimum abundance of magnetite is reached at the end of this interval. Interval 3 is marked by a progressive loss of hematite with depth, and at the base of this interval, 82% to 88% of the hematite component was depleted and the bulk magnetic mineral concentration was reduced to the lowest value in the entire studied mud section. Interval 4 has an increasing down-core enhancement of authigenic greigite, which is interpreted to have formed due to arrested pyritization resulting from consumption of pore water sulfate with depth. This is the first clear demonstration from an active depositional environment for a delay of thousands of years for acquisition of a magnetization carried by greigite. This detailed view of diagenetic processes in continental shelf sediments suggests that studies of geomagnetic field behavior from such sediments should be conducted with care. Paleoceanographic and paleoclimatic studies based on the magnetic properties of shelf sediments with high sedimentation rates like those in the Korea Strait are also unlikely to provide a meaningful signature associated with syn-depositional environmental processes. The rock magnetic properties of the surface sediments from the fine-grained depositional area on the outer shelf of the East China Sea, an area surrounded by sands, are investigated with a view to providing information on the sediment provenance. Multiple magnetic parameters such as magnetic susceptibility (%), anhysteretic remanent magnetization (ARM), saturation rernanent magnetization (SIRM), coercivities of SIRM (Her), and S ratios (relative abundance of low-coercivity magnetic minerals) are measured for all 179 surface samples, and partial representative samples are examined for their magnetic hysteresis parameters, temperature-dependence of magnetic susceptibility and x-ray diffraction spectra. Our research indicates that the magnetic mineralogy is dominated by magnetite with a small amount of hematite and is primarily of pseudo-single domain (PSD) to multidomain (MD) nature with a detrital origin. In the surface sediments, the granulometry of magnetic fractions is basically independent of grain sizes of the sediment containing the magnetic grains, and the composition of magnetic minerals remains almost homogeneous, that is, with a relatively constant ratio of low to high coercivity fraction throughout the area. The magnetic concentration in the study area generally decreases to the east or southeast accompanied by magnetic-particle fining to the east or to the northeast. The geographic pattern of magnetic properties is most reasonably explained by a major source of sediment jointly from the erosion of the old Huanghe River deposit and the discharge of the Changjiang River. The rock magnetic data facilitate understanding of the transport mechanism of fine-grained sediments in the outer shelf of the East China Sea.

西北太平洋某海域40ka来的地磁场相对强度研究

Paleointensity changes of geomagnetic field help us to understand the evolutionprocess of earth completely and provide further constraints for earth interior process and geodynamo model. Marine sediments are good carriers for relative paleointensity of geomagnetic field. But in most cases, deep sea sediments that conform with magnetic "uniformity" usually have low sedimentation rate about l-2cm/ka and lie under the Carbonate Compensate Depth with little carbonate content. Therefore, the number of relative paleointensity records with detailed oxygen stratigraphy is still rare. This thesis focus on four cores from east of Ryukyu Trench which have foraminiferal content of 5-30% and sedimentation rate of lOcm/ka and wish to get centennial -millennial changes of relative paleointensity.The sediments from east of Ryukyu Trench conform with magnetic "uniformity" and remanences of four cores all show single component with stable direction and faithfully record the magnetic field. The NRM301T1T/ARM and NRMsomT/ x are still affected by grain size and concentration changes although the sediments are "uniform" , indicating the uniformity might not enough for relative paleointensity. After renomalized by grain size parameter MDF, the intensities remove the effect of grain size changes to different degrees and show coherency in 1-1 Oka scale with results from ODP983/984. The characteristics of paleointensity of geomagnetic field arefrom 32-24kaBP, paleointensity of geomagnetic field is low;24-12kaBP, paleointensity of geomagnetic field is high and shows two peaks boundary with 19kaBP.3) 12-5.3ka, paleointensity is low. Then increase from 5,3kaBP until a small trough at 2.7kaBP, then increase till now.

Learning Fine Motion by Markov Mixtures of Experts

Compliant control is a standard method for performing fine manipulation tasks, like grasping and assembly, but it requires estimation of the state of contact between the robot arm and the objects involved. Here we present a method to learn a model of the movement from measured data. The method requires little or no prior knowledge and the resulting model explicitly estimates the state of contact. The current state of contact is viewed as the hidden state variable of a discrete HMM. The control dependent transition probabilities between states are modeled as parametrized functions of the measurement We show that their parameters can be estimated from measurements concurrently with the estimation of the parameters of the movement in each state of contact. The learning algorithm is a variant of the EM procedure. The E step is computed exactly; solving the M step exactly would require solving a set of coupled nonlinear algebraic equations in the parameters. Instead, gradient ascent is used to produce an increase in likelihood.

The Grasping Problem: Toward Task-Level Programming for an Articulated Hand

This report presents a system for generating a stable, feasible, and reachable grasp of a polyhedral object. A set of contact points on the object is found that can result in a stable grasp; a feasible grasp is found in which the robot contacts the object at those contact points; and a path is constructed from the initial configuration of the robot to the stable, feasible final grasp configuration. The algorithm described in the report is designed for the Salisbury hand mounted on a Puma 560 arm, but a similar approach could be used to develop grasping systems for other robots.

The Effect of Transmission Design on Force-Controlled Manipulator Performance

Previous research in force control has focused on the choice of appropriate servo implementation without corresponding regard to the choice of mechanical hardware. This report analyzes the effect of mechanical properties such as contact compliance, actuator-to-joint compliance, torque ripple, and highly nonlinear dry friction in the transmission mechanisms of a manipulator. A set of requisites for high performance then guides the development of mechanical-design and servo strategies for improved performance. A single-degree-of-freedom transmission testbed was constructed that confirms the predicted effect of Coulomb friction on robustness; design and construction of a cable-driven, four-degree-of- freedom, "whole-arm" manipulator illustrates the recommended design strategies.

Roles of Knowledge in Motor Learning

The goal of this thesis is to apply the computational approach to motor learning, i.e., describe the constraints that enable performance improvement with experience and also the constraints that must be satisfied by a motor learning system, describe what is being computed in order to achieve learning, and why it is being computed. The particular tasks used to assess motor learning are loaded and unloaded free arm movement, and the thesis includes work on rigid body load estimation, arm model estimation, optimal filtering for model parameter estimation, and trajectory learning from practice. Learning algorithms have been developed and implemented in the context of robot arm control. The thesis demonstrates some of the roles of knowledge in learning. Powerful generalizations can be made on the basis of knowledge of system structure, as is demonstrated in the load and arm model estimation algorithms. Improving the performance of parameter estimation algorithms used in learning involves knowledge of the measurement noise characteristics, as is shown in the derivation of optimal filters. Using trajectory errors to correct commands requires knowledge of how command errors are transformed into performance errors, i.e., an accurate model of the dynamics of the controlled system, as is demonstrated in the trajectory learning work. The performance demonstrated by the algorithms developed in this thesis should be compared with algorithms that use less knowledge, such as table based schemes to learn arm dynamics, previous single trajectory learning algorithms, and much of traditional adaptive control.