基于RBF神经网络的机械手轨迹规划及运动仿真研究-机械电子工程专业论文.docxVIP

ABSTRACT Robot’s trajectory planning is very important in the control of robotics．Based on demands and restrictions of AS—MRobot6 manipulator,this paper presents a novel method of trajectory planning．By the researching processes of concrete analysis of trajectory planning on robot’s manipulator aim，imitation of trajectory based on kinematics and optimization of trajectory in the articulation space，this paper establishes two models according to RBF neural networks algorithm，one is used for curve interpolation in operation space and the other is used for curve approximation in articulate space． To realize trajectory planning in operation space，all one—input-one。output RBF neural networks model is built and trained using x-axis value of discrete points as inputs and y-axis value outputs．Meanwhile．all one—input—six-output one is established and trained using discrete time as inputs and value of six angle outputs in articulate space． With character of rapid convergence and near approximation，this new algorithm is fault tolerant and irrelative with order of inputs，which can achieve firing requirement of engineering． Making use ofthe virtual prototyping technology,the manipulator’S model is founded in software ADAMS that developed for dynamics analysis of mechanical system．Furthermore， according to the planned path，the changes of the kinematics and dynamics indexes of the manipulator in the course of moving are studied and its effectiveness and feasibility validated． Key words：Manipulator,Dynamics，RBFNN，Prototype，Articulate space  独创性声明 本人声明所呈交的论文是我个人在导师指导下进行的研究工作及取得的研 究成果。尽我所知，除了文中特别加以标注和致谢的地方外，论文中不包含其他 人已经发表或撰写过的研究成果，也不包含为获得山东理工大学或其它教育机构 的学位或证书而使用过的材料。与我一同工作的同志对本研究所做的任何贡献均 已在论文中作了明确的说明并表示了谢意。 研究生签名 关于论文使用授权的说明 本人完全了解山东理工大学有关保留、使用学位论文的规定，即：学校有权 保留送交论文的复印件和磁盘，允许论文被查阅和借阅；学校可以用不同方式在 不同媒体上发表、传播学位论文的全部或部分内容，可以采用影印、缩印或扫描 等复制手段保存、汇编学位论文。 (必威体育官网网址的学位论文在解密后应遵守此协议) 研究生签名：孑，j鸯平 时间：≥。一7年6月毕日 引磁名：2涪号咖套∞7年绷岬  山东理工大学硕士学位论文 第一章绪论 第一章绪论 1．1模块