The Design And Implementation Of 6R Robot Simulation System

With the rise of the robotics industry, the demand of robots in various aspects including industry, agriculture and social life is growing, as well as the engineering and technical requirements of the robot, then how to develop efficient robots to meet the production needs is an urgent requirement.After studying the demands for the users of the simulation system, this paper conducts research on the mathematical model of 6R robot, achieving a 6R robot simulation system, mainly to complete the work in the following aspects:1. This paper choose to use the D-H method to set up the mathematical model of 6R robot and establish robot kinematics equations to study kinematics and inverse kinematics. Analyzing find that using the general algebraic method to inverse kinematics often occur no solution, this paper put to use the Newton iterative algorithm to improve this situation, and implement it in the simulation system. Comparing the system implement results and the experimental results using the general algebraic method find that Newton iterative algorithm can solve a lot of cases that using the general algebraic method will have no solution, not only greatly improve the stability of the system, but also lay a good foundation for the following trajectory planning;2. After conducting the study of 6R robot trajectory planning, this paper respectively research on the common polynomial interpolation algorithms in different trajectory planning space. The study find that although the polynomial interpolation algorithm can guarantee the robot end meet the kinematic constraints while moving according to the expected trajectory, but it’s computational complexity will continue to rise with the accuracy and interpolation constraints increasing, the algorithm efficiency is very low, and it is prone to "Runge shock". RBF neural network interpolation algorithm can solve the "Runge shock" problem and achieve better curve interpolation with a good nonlinearity. After research on the principle and processes of the RBF neural network algorithm, this paper implement it in the simulation system, and compare the results with the experimental results to polynomial interpolation algorithm. The comparison shows that the polynomial interpolation algorithm is unable to complete most of the trajectory planning with regular curve, while the RBF neural network algorithm used bythe system for robot trajectory planning is of good practicality and have a shorter response time;3. After analyzing system user’s needs, this paper use Matlab software together with associated toolkit to achieve a 6R robot simulation system with data load module, auxiliary tool modules and trajectory planning module. This paper complete the function that users can define the robot model and the work environment of the robot, the function that users can observe the robot in a three-dimensional perspective and converse the coordinate between the joint space and cartesian space accessibly, as well as the function that users can plan 6R robot trajectory in two preset condition. The the data loading module improve the usefulness of the system and the implement results of two preset condition together with their combinations essentially covers all possible trajectories, achieving a completeness simulation system. While analyzing the trajectory planning results of this system, it is found that this simulation system not only improves the efficiency and accuracy of trajectory planning and has a stable performance, but also can help users design robots with better performance.