Loading...

Multi-objective Design and Optimization of a Delta Robot and its Pareto-Optimal Solution and its Robust Control with Sliding Mode Method

Esmaeili Shahmiri, Yussuf Reza | 2018

468 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51326 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Vossoughi, Gholamreza; Moradi, Hamed
  7. Abstract:
  8. Nowadays, Automation and Robots have many industrial and non-industrial applications. One of the most important of Robotic applications in industries for speed increasing, is utilizing them for packing procedures on production lines. Speed and accuracy enhancement, low-error repeatability, and cost reduction are benefits if using robots. Delta robot, is one of the parallel robots which is consists of three arms each connected to universal joints at the base and can reach high speeds.Delta robot can be optimized in 1. Performance speed, 2.accuracy, 3. Maximum payload, 4. Workspace and dexterious workspace and 5. Costs reduction. Since it’s not possible to fully optimize all goals simultaneously, multi-optimized design will be applied to the robot and by utilizing Genetic Algorithm, Pareto front (non-dominant answers) will be resulted and design variable will be chosen with respect to it. Considered goals are minimizing mean power needed in motors, enhancing kinematic and dynamic performance, and avoiding serial and parallel singularities. Additionally, this project has been defined as another Master’s Thesis by Mr. Mehdi Sheikhzeinoddin and the goals and optimization results of both projects are effective on final results. Goals of this project are making the mass matrix near to diagonal, minimizing gravity effects and approaching the robot’s workspace to the desired one. (Robot’s size reduction). Design variables are first arm, second arm (parallelogram’s big arm), end effector radius and height of workpsace center from the base and will be numerically assigned with the results of both projects. After designing the robot, it should be controlled. Delta robot has a coupled and nonlinear dynamics and any movement of each arm, involves the others. Besides, since the robot must work accurately with the presence of noises, and frictional and inertial uncertainties, control design must be robust. Applied controlling method is the multi-layer robust sliding mode control. Sliding mode method can be designed to be robust and has less computational cost in comparison with other methods but for reducing errors, actuator chattering occurs. For solving this drawback, multi layers of errors are designed and error tends to zero layer by layer. Additionally, reducing correcting control amplitude with respect to the error, results in smoother input torques. Eventually, designed system will be simulated in software and final results are concluded for different system uncertainties
  9. Keywords:
  10. DELTA Robot ; Sliding Mode Control ; Parallel Robot ; Nonlinear Control ; Multiobjective Optimization ; Pareto Improving

 Digital Object List

 Bookmark

No TOC