【摘要】：連續(xù)型機器人是依據(jù)仿生學原理開發(fā)的新型機器人,在空間探測、核電維修、緊急救援等方面具有廣闊的應用前景,是目前機器人研究領域的熱點。機器人依靠仿生皮膚和仿生軀干實現(xiàn)運動,對于含有剛性脊柱的連續(xù)型機器人,其運動能力主要依靠仿生軀干提供。本文依據(jù)生物軀干的超冗余特性和生物彈性結(jié)構(gòu)開發(fā)出了可作為連續(xù)型機器人軀干的新型仿生機構(gòu),并通過動力學優(yōu)化和控制率設計,保證了其優(yōu)異的運動特性。利用仿生設計方法,該仿生機構(gòu)采用繩驅(qū)超冗余度機構(gòu),由六個串聯(lián)的萬向節(jié)和六組并聯(lián)彈簧組成,共12個自由度,整個模型由18根繩線驅(qū)動。在動力學模型中繩線和彈簧簡化為外界作用力,利用D-H參數(shù)描述機構(gòu)的位姿,通過力系簡化減少作用力數(shù)目,并求得繩線和彈簧作用力與等效關(guān)節(jié)力矩之間的雅可比矩陣。依據(jù)拉格朗日方程建立該機構(gòu)的動力學方程,并分析其動力學特性,驗證了機構(gòu)設計的合理性。仿生機構(gòu)的超冗余特性和冗余驅(qū)動特性,決定了其運動速度從操作空間向關(guān)節(jié)空間、驅(qū)動力從關(guān)節(jié)空間向驅(qū)動空間轉(zhuǎn)換時均存在多解的情況,因此運動學和動力學層面存在明顯的可優(yōu)化特性。使用待優(yōu)化目標,利用梯度法和零空間法可以選取合理的雅可比矩陣零空間向量,從而實現(xiàn)機構(gòu)在運動過程中的速度優(yōu)化和驅(qū)動力優(yōu)化。由于存在兩組欠定關(guān)系,該系統(tǒng)的可優(yōu)化能力較強,可以實現(xiàn)多目標優(yōu)化。控制系統(tǒng)用于維持機構(gòu)動作過程的穩(wěn)定性,現(xiàn)階段機器人普遍使用的控制方式為運動PD控制,不利于復雜機構(gòu)的運動實現(xiàn)。本文利用計算力矩控制和動力學優(yōu)化方法設計出了針對該連連續(xù)型冗余自由度機器人的優(yōu)化控制率,并利用李雅普諾夫函數(shù)證明了其穩(wěn)定性。仿真驗證表明,相比于運動PD控制,工作于該控制率下的系統(tǒng)具有較強的穩(wěn)定性和信號跟蹤能力,且能夠處理驅(qū)動力超限的問題。本文利用動力學模型和優(yōu)化控制率編寫控制程序,通過SIMULINK和ADAMS進行聯(lián)合仿真驗證了該機構(gòu)對于生物運動的模擬能力。仿真結(jié)果證明該機構(gòu)能夠很好地模擬機器人的平面運動和三維空間運動,同時具有極強的動力學優(yōu)化能力,適合作為連續(xù)型機器人的仿生脊柱。
[Abstract]:Continuous robot is a new kind of robot developed according to the principle of bionics. It has a broad application prospect in space exploration, nuclear power maintenance, emergency rescue and so on. It is a hot spot in the field of robot research at present. Robots rely on bionic skin and bionic torso to achieve motion. For continuous robots with rigid spine, their motion ability is mainly provided by bionic torso. In this paper, a new bionic mechanism which can be used as the torso of continuous robot is developed according to the super-redundancy characteristics of the biological trunk and the biological elastic structure. The excellent kinematic characteristics are ensured by dynamic optimization and control rate design. By using the bionic design method, the bionic mechanism is composed of six universal joints in series and six parallel springs, with 12 degrees of freedom. The whole model is driven by 18 rope lines. In the dynamic model, the rope line and spring are simplified as external forces, the position and orientation of the mechanism are described by D-H parameters, the number of forces is reduced by simplifying the force system, and the Jacobian matrix between the rope line and spring force and the equivalent joint torque is obtained. Based on the Lagrange equation, the dynamic equation of the mechanism is established, and its dynamic characteristics are analyzed. The rationality of the mechanism design is verified. The superredundancy and redundant driving characteristics of the bionic mechanism determine that there are multiple solutions when the motion speed changes from the operating space to the joint space, and the driving force changes from the joint space to the drive space. Therefore, kinematics and dynamics have obvious optimizable properties. Using the objective to be optimized, the gradient method and the zero space method can be used to select the reasonable Jacobian matrix null space vector, thus the velocity optimization and the driving force optimization of the mechanism in the course of motion can be realized. Due to the existence of two groups of undetermined relationships, the system has a strong ability to optimize and can realize multi-objective optimization. The control system is used to maintain the stability of the mechanism action process. At present, the control mode commonly used in robot is motion PD control, which is not conducive to the movement realization of complex mechanism. In this paper, the optimal control rate for the continuous redundant robot is designed by using the method of calculating torque control and dynamic optimization, and its stability is proved by using Lyapunov function. The simulation results show that the system working at this control rate has strong stability and signal tracking ability, and can deal with the problem of driving force exceeding the limit compared with the motion PD control. In this paper, the dynamic model and optimal control rate are used to compile the control program. The simulation results of SIMULINK and ADAMS show that the mechanism is capable of simulating biological motion. The simulation results show that the mechanism can well simulate the planar and three-dimensional motion of the robot, and has a strong dynamic optimization ability, which is suitable for the bionic spine of a continuous robot.
【學位授予單位】：哈爾濱工業(yè)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP242

【參考文獻】

相關(guān)期刊論文 前5條

1 翟士民;劉榮;薛彤;;繩驅(qū)動連續(xù)型機械臂設計[J];機械工程與自動化;2015年02期

2 何斌;王志鵬;唐海峰;;軟體機器人研究綜述[J];同濟大學學報(自然科學版);2014年10期

3 齊慶璽;黃攀峰;劉正雄;;冗余空間機器人基座姿態(tài)調(diào)整的笛卡爾路徑規(guī)劃[J];指揮控制與仿真;2014年01期

4 葉長龍;馬書根;李斌;王越超;;三維蛇形機器人巡視者Ⅱ的開發(fā)[J];機械工程學報;2009年05期

5 陳偉海,張啟先,楊宗煦,李健;冗余度機器人動力學優(yōu)化控制研究[J];北京航空航天大學學報;1998年05期

相關(guān)會議論文 前1條

1 郁樹梅;馬書根;李斌;王越超;;水陸兩棲蛇形機器人的上浮和下潛步態(tài)研究[A];第九屆全國信息獲取與處理學術(shù)會議論文集Ⅰ[C];2011年

相關(guān)博士學位論文 前3條

1 邵鐵鋒;氣動柔性象鼻型連續(xù)機器人研究[D];浙江工業(yè)大學;2014年

2 王鑫;類豹型四足機器人高速運動及其控制方法研究[D];哈爾濱工業(yè)大學;2013年

3 鄧奇;具彈性軀干仿生四足機器人準被動動力學研究[D];上海交通大學;2012年

相關(guān)碩士學位論文 前5條

1 劉天亮;面向狹小空間作業(yè)的繩索驅(qū)動超冗余機械臂的研究[D];哈爾濱工業(yè)大學;2016年

2 吳海波;具有可變剛度的四足機器人仿生脊柱設計與應用研究[D];北京交通大學;2016年

3 王超;線驅(qū)動硅膠軟體機械臂建模與控制[D];上海交通大學;2015年

4 佘宇;冗余機械臂單關(guān)節(jié)故障下的建模與容錯控制研究[D];哈爾濱工業(yè)大學;2013年

5 朱雷;自由浮動冗余度空間機器人運動規(guī)劃與力矩優(yōu)化方法研究[D];南京航空航天大學;2008年

，

本文編號：2396718