Date of Award
Mechanical, Automotive, and Materials Engineering
Articulated Robot, Artificial Neural Network, Inverse Kinematics, Machine Tool, Reconfigurable, Singularity
CC BY-NC-ND 4.0
Automation has led to industrial robots facilitating a wide array of high speed, endurance, and precision operations undertaken in the manufacturing industry today. An acceptable level of functioning and control is therefore vital to the efficacy and successful implementation of such manipulators. This research presents a comprehensive analytical tool for downstream optimization of manipulator design, functionality, and performance. The proposed model is reconfigurable and allows for modelling and validation of different industrial robots. Unique 3D visual models for a manipulator workspace and kinematic singularities are developed to gain an understanding into the task space and reach conditions of the manipulator's end-effector. The developed algorithm also presents a non-conventional and computationally inexpensive solution to the inverse kinematics problem through the use Artificial Neural Networks. Application of the proposed technique is further extended to aid in development of path planning models for a uniform, continuous, and singularity free motion.
Aggarwal, Luv, "Reconfigurable Validation Model for Identifying Kinematic Singularities and Reach Conditions for Articulated Robots and Machine Tools" (2014). Electronic Theses and Dissertations. 5219.