本文選題：機械臂　切入點：軌跡規(guī)劃　出處：《哈爾濱工業(yè)大學》2017年碩士論文

【摘要】：隨著國家《中國制造2025》的提出,將工業(yè)機器人等智能裝備作為未來發(fā)展的重點。對于工業(yè)級機械臂而言,軌跡規(guī)劃和軌跡跟蹤問題的研究為其能夠精準安全的工作提供了重要的保障。根據(jù)給定目標的具體任務可以準確的規(guī)劃出機械臂關節(jié)變量的運動軌跡,再對機械臂生成的軌跡進行軌跡跟蹤控制,使得控制輸入驅(qū)動力矩能夠達到機械臂精確跟蹤給定目標軌跡的需求。因此軌跡規(guī)劃和軌跡跟蹤是機械臂能夠完成給定任務的基礎,具有重要的研究意義。首先本文根據(jù)Denavit-Hartenberg法建立了BO-6-3六自由度機械臂的坐標模型并根據(jù)各個連桿的D-H參數(shù)求得了各個關節(jié)之間的位姿關系,進而可以通過齊次變換矩陣得到機械臂末端執(zhí)行器相對于機械臂本體的位姿,并推導出機械臂正運動學方程,分析了利用幾何法求取運動學逆解的簡要過程。采用拉格朗日公式法對平面內(nèi)兩自由度機械臂進行了機械臂動力學建模,同時分析了機械臂動力學正解和動力學逆解之間的關系。然后本文提出了基于差分進化(Differential Evolution)優(yōu)化BP神經(jīng)網(wǎng)絡求解機械臂運動學逆解的方法,并與BP神經(jīng)網(wǎng)絡進行了比較,仿真結(jié)果表明DEBP神經(jīng)網(wǎng)絡求得的逆解精度高同時也分析了傳統(tǒng)求解運動學逆解方法的不足之處。在關節(jié)空間和笛卡爾空間內(nèi)分別進行機械臂的軌跡規(guī)劃,在關節(jié)空間內(nèi)通過運動學的逆解求得關節(jié)角度值序列,并采用五次多項式插值法進行運算,求得了關節(jié)空間內(nèi)關節(jié)角的位置、速度和加速度的變化曲線。在笛卡爾空間內(nèi)采用直線插補法完成了從初始位置到終止位置的軌跡規(guī)劃,完成了目標指定任務。最后本文采用了雙冪次趨近律與改進的終端滑模面相結(jié)合的滑模變結(jié)構控制策略,對平面兩自由度機械臂進行軌跡跟蹤控制研究。針對傳統(tǒng)冪次趨近律收斂速度慢,抖振現(xiàn)象明顯等缺點,采用了雙冪次趨近律的滑模控制方法,保證了系統(tǒng)能夠在有限時間內(nèi)快速的到達滑動模面。與此同時傳統(tǒng)的終端滑模面在對機械臂關節(jié)角的位置誤差和速度誤差跟蹤時精度較低,也不能很好的控制當系統(tǒng)進入滑動模面瞬間的狀態(tài)情況,易于產(chǎn)生較強的抖振現(xiàn)象,因此本文又采用了改進的終端滑模面。將雙冪次趨近律和改進的終端滑模面結(jié)合后,針對機械臂動力學方程推導出機械臂系統(tǒng)的控制律。將在選定同一滑模面的情況下針對不同的趨近律進行對比分析,同理在選定同一趨近律的情況下針對不同的滑模面進行對比分析。仿真結(jié)果表明雙冪次趨近律和改進的終端滑模面的結(jié)合在具有外界干擾的情況下還能夠使機械臂系統(tǒng)保持較強的魯棒穩(wěn)定性和魯棒性,有效的提升了機械臂的跟蹤精度,并在一定程度上減弱了機械臂的抖振現(xiàn)象。
[Abstract]:With the proposal of "made in China 2025", intelligent equipment, such as industrial robots, will be regarded as the focus of future development.For industrial manipulator, the research on trajectory planning and trajectory tracking provides an important guarantee for its precise and safe work.According to the specific task of the given target, the trajectory of the manipulator joint variables can be accurately planned, and then the trajectory generated by the manipulator can be tracked and controlled.The control input drive torque can meet the requirements of the manipulator to track the given target trajectory accurately.Therefore, trajectory planning and trajectory tracking are the basis for the robot arm to complete a given task, which is of great significance.In this paper, the coordinate model of BO-6-3 six-degree-of-freedom manipulator is established according to the Denavit-Hartenberg method, and the position and pose relationship between joints is obtained according to D-H parameters of each connecting rod.Furthermore, the position and orientation of the end actuator relative to the body of the manipulator can be obtained by the homogeneous transformation matrix, and the forward kinematics equation of the manipulator is derived, and the process of obtaining the inverse kinematics solution by using the geometric method is analyzed.The dynamic modeling of the manipulator with two degrees of freedom in a plane is carried out by using the Lagrange formula method, and the relationship between the dynamic forward solution and the inverse solution of the manipulator is analyzed.Then, a BP neural network optimization method based on differential evolution evolution is proposed to solve the inverse kinematics of manipulator, and compared with BP neural network.The simulation results show that the inverse solution obtained by DEBP neural network has high accuracy and the shortcomings of the traditional inverse kinematics solution method are also analyzed.In the joint space and the Cartesian space, the trajectory planning of the manipulator is carried out, and the sequence of joint angle values is obtained by the inverse kinematics solution in the joint space, and the fifth order polynomial interpolation method is used to perform the operation.The position, velocity and acceleration curves of the joint angle in the joint space are obtained.In the Cartesian space, the trajectory planning from the initial position to the termination position is completed by using the linear interpolation method, and the target assignment task is completed.Finally, a sliding mode variable structure control strategy combining the double power approach law and the improved terminal sliding mode surface is used to study the trajectory tracking control of the planar two-degree-of-freedom manipulator.Aiming at the shortcomings of the traditional power law such as slow convergence rate and obvious chattering phenomenon, the sliding mode control method of the double power approach law is adopted to ensure the system can reach the sliding mode surface quickly in the limited time.At the same time, the precision of the traditional terminal sliding mode surface in tracking the position error and velocity error of the joint angle of the manipulator is low, and the system can not be controlled well when the system enters the instant state of the sliding mode surface, so it is easy to produce strong chattering phenomenon.Therefore, an improved terminal sliding mode surface is adopted in this paper.The control law of the manipulator system is derived based on the dynamic equation of the manipulator by combining the double power approach law with the improved terminal sliding mode surface.In the case of selecting the same sliding mode surface, the contrast analysis is carried out for different convergence laws, and the same law is chosen for different sliding mode surfaces.The simulation results show that the combination of the double power approach law and the improved terminal sliding mode surface can make the manipulator system keep robust stability and robustness, and improve the tracking accuracy of the manipulator effectively.The chattering phenomenon of the manipulator is reduced to some extent.
【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】：TP241

【參考文獻】

相關期刊論文 前10條

1 李慧潔;蔡遠利;;基于雙冪次趨近律的滑模控制方法[J];控制與決策;2016年03期

2 穆效江;;機械臂的自適應神經(jīng)滑�？刂品椒ㄑ芯縖J];控制工程;2015年04期

3 ;《中國制造2025》推動機器人發(fā)展[J];機器人技術與應用;2015年03期

4 王三秀;俞立;徐建明;邢科新;王正初;;機械臂自適應魯棒軌跡跟蹤控制[J];控制工程;2015年02期

5 孫英飛;羅愛華;;我國工業(yè)機器人發(fā)展研究[J];科學技術與工程;2012年12期

6 王鴻鵬;楊云;劉景泰;;高速移動機器人的研究現(xiàn)狀與發(fā)展趨勢[J];自動化與儀表;2011年12期

7 畢昆;趙馨;侯瑞鋒;王成;;機器人技術在農(nóng)業(yè)中的應用方向和發(fā)展趨勢[J];中國農(nóng)學通報;2011年04期

8 嵇鵬程;沈惠平;;服務機器人的現(xiàn)狀及其發(fā)展趨勢[J];常州大學學報(自然科學版);2010年02期

9 榮盤祥;楊晶;胡林果;馬廣富;;基于RBF網(wǎng)絡的SCARA機器人的運動學逆解[J];電機與控制學報;2007年03期

10 杜慧秋;基于模糊自適應不確定性機械臂的軌跡跟蹤控制[J];電機與控制學報;2005年03期

相關博士學位論文 前1條

1 梅紅;移動機械手的逆運動學及滑模變結(jié)構軌跡跟蹤控制研究[D];山東大學;2009年

相關碩士學位論文 前2條

1 李浩;六自由度機械臂軌跡規(guī)劃方法研究[D];浙江工業(yè)大學;2015年

2 安育紅;氣動機械手關節(jié)的研究[D];浙江大學;2007年

，

本文編號：1711213