6R串联机械臂复杂空间环境路径规划研究

发布时间：2018-06-25 13:53

本文选题：六自由度串联机械臂 + 运动学分析　；参考：《济南大学》2017年硕士论文

【摘要】：随着人工智能和自动化技术的不断发展,传统的机械臂进行重复动作作业已经不能符合时代的潮流。在当前的科技浪潮下,需要机械臂具备更高的智能性和自主性。本文基于RBT-6T/S03S型六自由度串联机械臂,进行了机械臂在静态环境下的避障路径规划研究。本文所做的研究工作包括六自由度串联机械臂运动学模型建立、正逆运动学求解、基于改进RRT算法和人工势场法的避障路径规划以及实验验证,各部分承上启下。主要内容如下:(1)运动学模型的建立:机械臂运动学模型是进行机械臂避障路径规划研究的基础,本文首先给出了机械臂空间位姿的数学表示方法,然后研究了机械臂空间的坐标变换,最后根据机械臂本体的具体数学参数,建立了机械臂的D-H坐标系,给出机械臂的连杆变化矩阵。(2)正逆运动学求解:在机械臂D-H坐标系下,结合机械臂的连杆变化矩阵,先给出了机械臂的正运动学方程,然后分别基于解析法和RBF神经网络,给出了机械臂逆运动学的求解方法。(3)避障路径规划:结合上面所完成的工作,基于传统的单枝RRT算法,从笛卡尔空间规划出发,结合机械臂的运动学约束,将其从二维平面扩展到三维空间,进行机械臂的避障路径规划。随后,采用bi-RRT对RRT算法进行了改进。最后,再从关节空间规划出发,采用势函数代替传统人工势场法中的合力作用,进行了机械臂的空间避障路径规划。(4)实验验证:分别设计了正运动学实验、逆运动学实验、基于RRT算法的路径规划实验和基于人工势场法的路径规划实验,对本文所做的工作进行实验验证。通过实验分析,本文所做的研究工作具有一定实际应用价值和参考意义,具有可行性和有效性。
[Abstract]:With the continuous development of artificial intelligence and automation technology, the traditional manipulators are unable to carry out repetitive actions in line with the trend of the times. In the current wave of science and technology, the robot arm needs a higher degree of intelligence and autonomy. Based on RBT-6T / S03S six-degree-of-freedom series manipulator, the obstacle avoidance path planning of the manipulator in static environment is studied in this paper. The research work in this paper includes the establishment of kinematics model of six-degree-of-freedom series manipulator, the solution of forward and inverse kinematics, the obstacle avoidance path planning based on improved RRT algorithm and artificial potential field method, and the experimental verification. The main contents are as follows: (1) Establishment of kinematics model: kinematics model of manipulator is the basis of studying obstacle avoidance path planning. Then the coordinate transformation of the manipulator space is studied. Finally, according to the specific mathematical parameters of the manipulator body, the D-H coordinate system of the manipulator is established, and the linkage change matrix of the manipulator is given. (2) the forward and inverse kinematics solution: in the D-H coordinate system of the manipulator, The forward kinematics equations of the manipulator are given based on the change matrix of the connecting rod, and then based on the analytical method and the RBF neural network, respectively, the forward kinematics equations of the manipulator are given. The inverse kinematics method of the manipulator is presented. (3) obstacle avoidance path planning: based on the traditional single-branch RRT algorithm, the kinematics constraints of the manipulator are combined with the Cartesian space planning. It is extended from two-dimensional plane to three-dimensional space to plan the obstacle avoidance path of the manipulator. Then, the RRT algorithm is improved by bi-RRT. Finally, starting from joint space planning, potential function is used to replace the resultant force in the traditional artificial potential field method, and the space obstacle avoidance path planning of the manipulator is carried out. (4) Experimental verification: the forward kinematics experiment and the inverse kinematics experiment are designed respectively. The path planning experiment based on RRT algorithm and the path planning experiment based on artificial potential field method are carried out to verify the work done in this paper. Through experimental analysis, the research done in this paper has some practical application value and reference significance, and is feasible and effective.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP241

【参考文献】