胸鳍推进机器鱼俯仰稳定性分析

胸鳍推进机器鱼的胸鳍在通过扑动产生推力的同时,也会产生纵向扰动,引起机身的俯仰运动,对应用产生一系列的影响.本文在传统潜艇稳定性概念的基础上建立了胸鳍推进机器鱼的俯仰稳定性模型,证明了机器鱼的俯仰运动系统是一个二阶线性系统;分析了由胸鳍扑动引入的扰动,证明其中包含常量分量、扑动频率的一倍频分量和二倍频分量.最后进行了样机的游动实验,实验结果间接证明了俯仰运动模型和胸鳍扑动扰动模型的正确性.     

仿生机器鱼胸/尾鳍协同推进闭环深度控制

为改善机器鱼定深控制过程中的动态性能与稳态性能,根据深度误差的大小将定深控制过程分解为趋近阶段与巡游阶段,给出了一种基于中枢模式发生器(central pattern generator,CPG)与模糊控制相结合的闭环运动控制方法.为此,首先建立了以压力传感器信号为反馈输入,通过模糊控制器调节控制参数的CPG运动控制模型.在此基础上,针对误差较大的趋近阶段,采用胸/尾鳍协同方式,通过趋近模糊控制器改变摇翼关节的偏置量与幅值来使机器鱼快速到达期望深度;针对误差较小的巡游阶段,采用改变攻角方式,通过巡游模糊控制器改变胸鳍攻角来使机器鱼保持在期望深度.两阶段之间通过胸鳍CPG的启停实现切换.模糊控制器设计时利用了基于最小二乘法对实验数据拟合而得出的俯仰角变化率与控制参数的近似关系,提高了机器鱼趋向期望深度的速度并减小了在期望深度巡游时的稳态误差.仿真与实验结果验证了所提控制方法的有效性.     

基于模糊控制的仿生机器鱼转向控制研究

针对一类微小型仿生机器鱼的转向控制问题,结合其自身运动的特点,提出一种基于模糊控制的转向控制算法.根据实验室经验建立模糊控制规则,运用Mamdani推理,构造出控制响应表.通过避障问题对算法进行验证,仿真结果表明,该方法对机器鱼的转向控制有效,可行且能满足实时性的要求.     

仿生机器鱼技术研究进展及关键问题探讨

本文简要介绍了仿生机器鱼技术在仿鱼推进机理研究中的地位和作用,探讨了这一领域关键的研究课题,并介绍了国内外几个机器鱼平台的研究进展。     

仿生机器鱼平台研究

本项目综合运用机械设计、电子电路、控制理论、波动推进理论、微处理器技术、传感器技术以及无线通信技术等多学科知识,基于ATMEL公司AVR系列单片机（ATmega128）,采用模块化设计,研制出一款密封性良好的具有尾部推进模块、升潜控制模块、自主避障模块、无线遥控模块、超声波测距模块、图像识别模块的波动推进式两关节仿生机器鱼水下作业平台,并对其直线推进、转弯、升潜、自主避障、无线控制功能进行了水下试验。论文从机械结构、外围电路、控制电路、核心功能实现四个主要方面进行论述。     

多仿生机器鱼控制与协调

仿生机器鱼是包含水动力学控制和机器人技术的多学科问题。本文从机器鱼的运动学模型人手,解决运动控制问题;以实时视觉信息处理技术为前提,为多鱼协调定位提供位姿信息;基于行为选择机制的协作策略,为机器鱼的任务级协作打下基础;最后,将所有算法集成到我们开发的多仿生机器鱼协调系统(Multiple Robot Fishes corrdinate System,MRFS)上。     

水下仿生机器鱼的研究进展Ⅳ——多仿生机器鱼协调控制研究

本文探讨了多仿生机器鱼群体游动协调的问题,这是仿生机器鱼技术研究的一个新的方向,目前尚无人涉足.首先研制了一套具有高效、高机动性的微小型多机器鱼平台,提出了基于AGENT的网格算法进行多机器鱼的定位和协调控制,利用该实验平台,进行了多机器鱼对抗和多机器鱼协调过孔的实验研究.     

水下仿生机器鱼的研究进展IV——多仿生机器鱼协调控制研究

本文探讨了多仿生机器鱼群体游动协调的问题,这是仿生机器鱼技术研究的一个新的方向,目前尚无人涉足．首先研制了一套具有高效、高机动性的微小型多机器鱼平台,提出了基于AGENT的网格算法进行多机器鱼的定位和协调控制,利用该实验平台,进行了多机器鱼对抗和多机器鱼协调过孔的实验研究．     

仿生机器鱼的检测与特征提取

在现有仿生机器鱼平台的基础上研究了仿生机器鱼的检测和特征提取算法,并采用均值背景模型结合大津法计算的阈值计算出机器鱼位置,采用前后背景的自适应更新方式减少光照改变的影响以及机器鱼影子的干扰。然后通过数学形态学处理,得到最终的二值图像并将符合要求的二值图像块数目标记在左上角。将目标块用外接矩形围住,通过相应的公式计算出各机器鱼二值图像的几何特征。实验结果证明可以达到预期要求。     

微小型仿生机器鱼设计与实时路径规划

提出一种结构紧凑, 运动灵活, 装配多传感器, 可自主游动的微小型仿生机器鱼系统设计方案. 在仿鲹科加月牙尾模式鱼类运动研究的基础上, 给出了微小型机器鱼推进、转弯等运动的控制方法. 结合运动控制和传感器信息处理, 给出了基于红外传感器的自主避障算法和基于光敏传感器的主动趋光算法, 进而提出了基于这两种传感器信息感知的动态光源跟踪方法. 通过实验, 给出了机器鱼尾部摆动频率、幅度和运动速度之间的关系, 验证了机器鱼追踪动态光源算法, 表明了本文所提系统设计方案和算法的有效性.     

仿生机器鱼运动控制算法研究

footnotesize A practical motion control strategy for a radio-controlled, 4-link and free-swimming biomimetic robot fish is presented. Based on control performance of the fish the fish's motion control task is decomposed into on-line speed control and orientation control. The speed control algorithm is implemented by using piecewise control, and orientation control is realized by fuzzy logic. Combining with step control and fuzzy control, a point-to-point (PTP) control algorithm is proposed and applied to the closed-loop experimental system that uses a vision-based position sensing subsystem to provide feedback. Experiments confirm the reliability and effectiveness of the presented algorithms.     

3-D Locomotion control for a biomimetic robot fish

This paper concerns with 3-D locomotion control methods for a biomimetic robot fish. The system architecture of the fish is firstly presented based on a physical model of carangiform fish. The robot fish has a flexible body, a rigid caudal fin and a pair of pectoral tins, driven by several servomotors. The motion control of the robot fish are then divided into speed control, orientation control, submerge control and transient motion control, corresponding algorithms are detailed respectively. Finally, experiments and analyses on a 4-link, radio-controlled robot fish prototype with 3-D locomotion show its good performance.     

仿鲹科机器鱼的倒退游动控制

给出了一种新型的仿鲹科机器鱼倒退游动控制方法. 在已有多关节仿鲹科机器鱼的基础上, 仿照欧洲鳗鱼的倒退运动机理, 修正机器鱼尾部关节的摆动规律, 进而实现倒游运动. 给出鲹科机器鱼游动的定性分析, 用于分析仿鲹科机器鱼推进机理和倒退游动的方法. 组合其他运动, 实现倒退游动、倒游中转弯等运动. 通过实验, 给出了机器鱼摆动频率和倒退运动速度之间的关系, 验证了本文所提方法的有效性.     

A fish robot driven by piezoceramic actuators and a miniaturized power supply

In this paper, we have introduced a prototype of a fish robot driven by unimorph piezoceramic actuators. To improve the swimming performance of the fish robot in terms of tail-beat angle, swimming speed, and thrust force, we used four light-weight piezo-composite actuators (LIPCAs) instead of the two LIPCAs used in the previous model. We also developed a new actuation mechanism consisting of links and gears. Performance tests of the fish robot were conducted in water at various tail-beat frequencies to measure the tail-beat angle, swimming speed, and thrust force. The tail-beat angle was significantly better than that of the previous model. The best tail-beat frequency of the fish robot was 1.4 Hz and the maximum thrust force was 0.0048 N. A miniaturized power supply, which was developed to excite the LIPCAs, was installed inside the fish robot body for free swimming. The maximum free-swimming speed was 3.2 cm/s. Recommended by Editorial Board member Hyoukryeol Choi under the direction of Editor Jae-Bok Song. This work was supported by the Korea Research Foundation under grant KRF-2004-005-D00045. Quang Sang Nguyen received the BS (2001) and MS (2006) from Hochiminh City University of Technology, Vietnam. Formerly an assistant lecturer of Naval Architect and Marine Engineering, Hochiminh City University of Technology, Vietnam (2001-2006), he is currently a Ph.D. student at the Department of Advanced Technology Fusion, Konkuk University. His specialty is biomimetic system design and smart material application. Seok Heo received the B.S., M.S. and Ph.D. degrees in Mechanical Engineering from Dongguk University in 1998, 2000, and 2003, Respectively. Currently he is a Research Professor at the Artificial Muscle Researcch Center, Konkuk University, Seoul, Korea. His research interests include biomimetics, vibration analysis, system design and control, and smart materials and structures. Hoon Cheol Park received the B.S. (1985) and M.S. (1987) from Seoul National University in Seoul, Korea and Ph.D. (1994) from the University of Maryland at College Park, MD, USA. He joined the Department of Aerospace Engineering, Konkuk University in Seoul, Korea, in 1995, and he is currently a Professor in the Department of Advanced Technology Fusion. His professional experience includes Kia Motors (1986–1988) and Korea Aerospace Research Institute (1994–1995). His specialty is finite element analysis and his recent research has focused on biomimetics. Nam Seo Goo graduated with honors in 1990 from the Department of Aeronautics Engineering of Seoul National University, and he got a masters degree and Ph.D. from the Department of Aerospace Engineering at the same university in 1992 and 1996, respectively. His Ph.D. thesis was on the structural dynamics of aerospace systems. As soon as he obtained the Ph.D. he entered the Agency for Defense development as a senior researcher. In 2002, after four years of service, he joined the Department of Aerospace Engineering at Konkuk University, Seoul, Korea, where he is currently serving as an Associate Professor of the Department of Advanced Technology Fusion. His current research interests include structural dynamics of small systems, smart structures and materials, and MEMS applications. Taesam Kang is a Professor of the Department of Aerospace and Information System Engineering, Konkuk University. He received the B.S., M.S. and Ph.D. degrees from Seoul National University in 1986, 1988 and 1992, respectively. His current research areas are robust control theories and the application of those theories with regard to flight control, development of micro-aerial vehicles and fish robots. Kwang Joon Yoon was awarded the BS (1981) and M.S. (1983) in Aeronautics Engineering from Seoul National University and Ph.D. (1990) in Aeronautics and Astronautics Engineering from Purdue. Since 1991 he has been a Professor at Konkuk University in Korea, where he is currently a Professor of Aerospace Engineering, the Director of the National Research Laboratory for Active Structures and Materials, the Director of the Artificial Muscle Research Center, and the Director of the Smart Robot Center. His current research interests include smart structures and materials, micro-aerial vehicles, and insect-mimetic micro-robot systems. Seung Sik Lee received the B.S. (1996) and M.S. (1998) in Civil Engineering from Hongik University in Seoul, Korea and Ph.D. (2003) in Civil Engineering from Georgia Institute of Technology, GA, USA. Currently he is a Senior Researcher at Korea Institute of Marine Science & Technology Promotion.     

两足步行机器人动态步行姿态稳定性及姿态控制

本文提出了一种新型两足步行机器人动态步行实时控制方法，引入了姿态稳定性和步态稳定性的概念，把两足动态步行控制分为姿态稳定性控制和步态稳定性控制两个相互联系的子系统，并从姿态稳定性分析出发，重点研究了各关节轨迹跟踪控制和系统整体动态平衡等问题，提出了姿态控制器的结构及设计方法．     

Modeling and simulation of an artificial muscle and its application to biomimetic robot posture control

The shape memory effect exhibited by Nitinol wire can be utilized to construct an artificial muscle. The muscle is activated by an electric current, which produces heat and initiates a phase transformation. The Nitinol artificial muscle stress–strain–power relationship was determined by experiments, and a mathematical model was developed. The artificial muscle model was utilized for the posture control of a biomimetic underwater robot. The optimal activation patterns for height, pitch, and roll postures were determined. Simulation results for the height postures are in agreement with the experiments. The separation between the center of gravity and the centroid of the robot has a stabilizing effect on pitch and roll postures.     

Undulatory locomotion and effective propulsion for fish-inspired robot

Swimming, turning, and whip-sweeping propulsion for carangiform locomotion of a fish robot are investigated by means of a 4-link planar tail and an autonomous underwater vehicle (AUV)-like model. It is observed that excellent acceleration occurs when a whip sweeping behavior has been applied to the fish tail. The forward speed can even increases twice to the nominal swimming via the simulation study. The efficient movement is thus incorporated to the fish robot for agile movement. The robot's swimming patterns realize the effect in terms of the forward swimming, turning swimming, acceleration increasing, descended swimming, ascended swimming, depth regulating, and self-stabilization. Verification is accomplished by incorporating the 4-link planar tail, AUV-like model, and a two degree-of-freedom (DOF) barycenter mechanism. The four-link planar tail and 2-DOF barycenter mechanism act, respectively, as the thrust generator and stabilizing actuator for the fish robot. Sliding mode control (SMC) has been applied for three-dimensional (3D) trajectory tracking. Simulation results illustrate satisfactory performances of the fish robot in terms of the fish-like behaviors and maneuverability, which are due to the consequence of the mimicked predator-fish behaviors and performance robustness of the SMC for trajectory tracking under ocean current perturbations and modeling uncertainties.     

仿人型机器人动态步行控制方法

本文介绍了仿人型机器人动态步行的一些基本问题和相关概念．从信息和控制的角度对近年来仿人型机器人动态步行研究中出现的步态规划和姿态控制方法进行了分析,并指出了它们的特点．提出了先进仿人型机器人实现过程中值得进一步研究的问题．     

点接触式两足步行机器人直立姿态稳定控制

针对点接触式两足步行机器人提出了一种直立姿态稳定控制策略。建立了机器人动力学模型,对机器人直立姿态的稳定性和可控性进行分析;并基于T-S模型建模,构造机器人全局模糊模型,基于LQR最优控制理论,设计全局渐近稳定的模糊控制器对机器人直立姿态进行稳定控制。仿真结果表明,该算法具有较宽的稳定范围,可以使机器人在上电以后和从行走切换到站立状态时迅速恢复直立姿态,保持稳定。