【摘要】：我國是一個(gè)制造業(yè)大國,要向著制造業(yè)強(qiáng)國的方向轉(zhuǎn)變,工業(yè)機(jī)器人在其中發(fā)揮著非常重要的作用。在機(jī)械臂的研究中,運(yùn)動(dòng)學(xué)是基礎(chǔ),運(yùn)動(dòng)規(guī)劃是期望軌跡來源,軌跡跟蹤控制是核心。本文基于這三個(gè)方面做出了相應(yīng)研究。(1)機(jī)械臂的運(yùn)動(dòng)學(xué):首先,以Denso VP6242G機(jī)器人為對象,推導(dǎo)出正運(yùn)動(dòng)學(xué)方程,根據(jù)正運(yùn)動(dòng)學(xué)方程用傳統(tǒng)的反變換法解出運(yùn)動(dòng)學(xué)逆解,并仿真驗(yàn)證了運(yùn)動(dòng)學(xué)正逆解的正確性;然后,針對傳統(tǒng)方法對一般結(jié)構(gòu)的機(jī)械臂運(yùn)動(dòng)學(xué)逆解無法求解的問題,研究了一種適用于任何結(jié)構(gòu)的不含反三角的高精度實(shí)時(shí)求解運(yùn)動(dòng)學(xué)逆解的解法,該解法基于改進(jìn)的萬有引力粒子群算法(PSOGSA)。通過仿真驗(yàn)證了該解法的準(zhǔn)確性、快速性和求解唯一性;最后,在工具坐標(biāo)系標(biāo)定中,針對裝有工具的機(jī)械臂標(biāo)定困難的偏差大的問題,在多點(diǎn)多姿態(tài)估算法的基礎(chǔ)上研究了一種新的快捷標(biāo)定方法并加入判誤機(jī)制,通過實(shí)驗(yàn)驗(yàn)證了所研究算法的正確性。(2)機(jī)械臂的運(yùn)動(dòng)規(guī)劃:在總結(jié)一般形式運(yùn)動(dòng)規(guī)劃的基礎(chǔ)上,探討用人工勢場法自動(dòng)避障的路徑規(guī)劃,并對該方法進(jìn)行了簡單避障仿真。同時(shí),研究了一種5-3-5樣條曲線插值的關(guān)節(jié)軌跡規(guī)劃方法,該方法在保障速度、加速度連續(xù)可導(dǎo)的前提下,還能保障加加速度連續(xù)且起止于零,比常規(guī)方法更有利于抑制機(jī)械臂的振動(dòng),通過Matlab仿真和Denso機(jī)械臂平臺(tái)驗(yàn)證了該方法的有效性。(3)機(jī)械臂的軌跡跟蹤控制:通過簡化Puma機(jī)械臂的六軸動(dòng)力學(xué)模型得到后三軸鎖死的前三軸模型,并在總結(jié)分析常用關(guān)節(jié)期望軌跡生成方法的基礎(chǔ)上,得到機(jī)械臂關(guān)節(jié)軌跡跟蹤控制的通用結(jié)構(gòu)。同時(shí),研究了PD、基于重力補(bǔ)償?shù)腜D、基于計(jì)算力矩補(bǔ)償?shù)腜D等關(guān)節(jié)軌跡跟蹤控制方法,在此基礎(chǔ)上研究了時(shí)變PD參數(shù)的計(jì)算力矩神經(jīng)網(wǎng)絡(luò)補(bǔ)償控制方法,有效改善了綜合控制效果。
[Abstract]:China is a big manufacturing country. Industrial robots play a very important role in the direction of changing to a powerful manufacturing country. In the research of manipulator, kinematics is the foundation, motion planning is the expected trajectory source, trajectory tracking control is the core. This paper makes corresponding research based on these three aspects. (1) Kinematics of the manipulator: firstly, taking the Denso VP6242G robot as the object, the forward kinematics equation is deduced, according to the forward kinematics equation, the inverse kinematics solution is solved by the traditional inverse transformation method, and the kinematics inverse solution is obtained by using the traditional inverse transformation method according to the forward kinematics equation. The correctness of forward and inverse kinematics solution is verified by simulation. Then, aiming at the problem that the traditional method can not solve the inverse kinematics solution of the mechanical arm of the general structure, a high-precision real-time solution of inverse kinematics solution without anti-trigonometry which is suitable for any structure is studied. The solution is based on an improved gravitational particle swarm optimization algorithm (PSOGSA). The accuracy, rapidity and uniqueness of the solution are verified by simulation. Finally, in the calibration of tool coordinate system, aiming at the difficulty of calibrating manipulator with tools, a new fast calibration method is studied on the basis of multi-point and multi-attitude estimation algorithm, and the mechanism of error judgment is added. The correctness of the proposed algorithm is verified by experiments. (2) Motion planning of manipulator: on the basis of summarizing the general form of motion planning, the path planning of automatic obstacle avoidance by artificial potential field method is discussed. A simple obstacle avoidance simulation is carried out. At the same time, a joint trajectory planning method based on the interpolation of 5 ~ 3 ~ 5 spline curves is studied. This method can guarantee continuous acceleration and zero acceleration under the premise that velocity and acceleration can be continuously derivable. It's better than conventional methods to suppress the vibration of the manipulator, The effectiveness of the proposed method is verified by Matlab simulation and Denso manipulator platform. (3) trajectory tracking control of the manipulator: by simplifying the six-axis dynamic model of the Puma manipulator, the front triaxial model of the post-triaxial locking is obtained. On the basis of summarizing and analyzing the common methods of joint expected trajectory generation, the general structure of manipulator joint trajectory tracking control is obtained. At the same time, the PD,-based gravity compensation PD, based on the calculated moment compensation of the joint trajectory tracking control method PD, on the basis of which the time-varying PD parameters of the calculated moment neural network compensation control method is studied. The comprehensive control effect is improved effectively.
【學(xué)位授予單位】：湖北工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP242

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王淑青;王亞洲;許琛;潘健;;改進(jìn)粒子群算法在機(jī)器人位置逆解上的應(yīng)用[J];湖北工業(yè)大學(xué)學(xué)報(bào);2017年01期

2 鐘斌;;不確定關(guān)節(jié)機(jī)器人模型的神經(jīng)網(wǎng)絡(luò)補(bǔ)償自適應(yīng)控制[J];機(jī)械科學(xué)與技術(shù);2017年03期

3 趙皓;劉滿祿;張華;;基于三次非均勻B樣條的6DOF機(jī)械臂軌跡規(guī)劃[J];機(jī)器人技術(shù)與應(yīng)用;2016年06期

4 張道德;徐波;;機(jī)器人多段插值軌跡規(guī)劃研究與分析[J];湖北工業(yè)大學(xué)學(xué)報(bào);2016年05期

5 夏正亞;洪亮;;基于多項(xiàng)式擬合插值函數(shù)的碼垛機(jī)器人軌跡規(guī)劃[J];山東科學(xué);2016年05期

6 倪天偉;江紅;林金珠;;基于改進(jìn)人工勢場法的移動(dòng)機(jī)器人避障路徑規(guī)劃算法[J];常州大學(xué)學(xué)報(bào)(自然科學(xué)版);2016年05期

7 張?jiān)品?馬振書;孫華剛;陸繼山;;基于改進(jìn)快速擴(kuò)展隨機(jī)樹的機(jī)械臂路徑規(guī)劃[J];火力與指揮控制;2016年05期

8 李光;符浩;;無模型機(jī)械臂BP神經(jīng)網(wǎng)絡(luò)狀態(tài)觀測及反演跟蹤控制[J];中國機(jī)械工程;2016年07期

9 陳鵬;李洪誼;;FS-SEA柔性臂改進(jìn)的反饋計(jì)算力矩控制方法[J];載人航天;2016年02期

10 齊飛;平雪良;劉潔;蔣毅;;關(guān)于工業(yè)機(jī)器人標(biāo)定方法的研究[J];機(jī)床與液壓;2015年21期

相關(guān)碩士學(xué)位論文 前10條

1 何婷婷;基于免疫多目標(biāo)的六自由度機(jī)械臂軌跡規(guī)劃[D];寧波大學(xué);2015年

2 杜媛媛;基于迭代預(yù)測控制的機(jī)械臂軌跡跟蹤的研究[D];浙江工業(yè)大學(xué);2014年

3 張翠;基于RBF神經(jīng)網(wǎng)絡(luò)的機(jī)械臂運(yùn)動(dòng)控制算法及應(yīng)用研究[D];蘭州交通大學(xué);2014年

4 張秀林;基于遺傳算法的機(jī)械臂時(shí)間最優(yōu)軌跡規(guī)劃[D];蘭州理工大學(xué);2014年

5 龐飛;工業(yè)機(jī)器人二次最優(yōu)軌跡規(guī)劃算法研究[D];東北大學(xué);2014年

6 李小倩;基于滑模的剛性機(jī)械臂有限時(shí)間軌跡跟蹤控制研究[D];東北大學(xué);2013年

7 蘭文寶;基于PMAC的工業(yè)機(jī)器人控制系統(tǒng)研究與實(shí)現(xiàn)[D];哈爾濱工程大學(xué);2013年

8 胡傳俊;神經(jīng)網(wǎng)絡(luò)在機(jī)械手逆解求解中的應(yīng)用[D];湘潭大學(xué);2011年

9 梁飛;基于虛擬現(xiàn)實(shí)技術(shù)的六自由度機(jī)械臂視景仿真與碰撞檢測系統(tǒng)[D];西華大學(xué);2011年

10 馬培羽;六自由度機(jī)械臂避障路徑規(guī)劃研究[D];華南理工大學(xué);2010年

，

本文編號：2467238