Robot Control

The goal of this work is the development of control concepts for a high precision robot joint. These proposed concepts are used to overcome the negative effects of the gear transmission. Thereby, the lowest achievable dynamic positioning error for the given setup is identified. For this purpose, a mechanical setup was used, in which the torque is generated by a permanent magnet synchron machine (PMSM) and transferred to the load through a Harmonic Drive gear. Two angular encoders on each side of the Harmonic Drive gear are supplemented by a Ferraris acceleration sensor on the load side.

Within this work a detailed model of the system, including the relevant non-linearities, is derived using Matlab/Simulink. The simulation results are verified using the described hardware and the model is refined in several iterations. This allows to gain a better understanding of the existing effects and therefore to develop appropriate concepts to increase the dynamic positioning precision. Finally, the control algorithms were implemented on the servo drive as a C# real time application. Beside a conventional cascade controller structure, a so called active damping for the suppression of resonant oscillations, as well as a compensation of the gear transmission error have been implemented. By this means a maximum load position error of 150 μrad could be achieved.

This paper describes the MATLAB/SIMULINK realization of the PUMA 560 robot manipulator
position control methodology. This paper focuses on two main areas, namely robot manipulator
analysis and implementation, and design, analyzed and implement nonlinear sliding mode control
(SMC) methods. These simulation models are developed as a part of a software laboratory to
support and enhance graduate/undergraduate robotics courses, nonlinear control courses and
MATLAB/SIMULINK courses at research and development company (SSP Co.) research center,
Shiraz, Iran.

The focus of this research is on the development, modeling and high precision robust control of an electro-mechanical continuum robot manipulator that serves as a sensing and motion system for hybrid testing. In this research parallel fuzzy logic theory is used to compensate the system dynamic uncertainty controller based on sliding mode theory. This design resulted in strongly non-linear and coupled dynamics as well as an inertial moving platform that attracted model-based control strategies. A novel non-linear control technique based on sliding mode Lyapunov based was selected to meet the multiple simultaneous specification control of nonlinear, uncertain and asymptotic tracking. Sliding mode controller (SMC) is a significant nonlinear controller under condition of partly uncertain dynamic parameters of system. This controller is used to control of highly nonlinear systems especially for continuum robot manipulator, because this controller is robust and stable in presence of partly uncertainties. Sliding mode controller was used to achieve a stable tracking, while the parallel fuzzy-logic optimization added intelligence to the control system through an automatic tuning of the sliding mode methodology uncertainties. Simulation results demonstrated the validity of the Mamdani parallel fuzzy-optimization control with asymptotic and stable tracking at different position inputs. This compensation demonstrated a well synchronized control signal at different excitation conditions.

This contributed research book contains eight chapters that present important aspects of robot motion and control. All principles, models and methods are field tested and can be readily used for solving real-world problems, such as factory automation, disposal of nuclear wastes, landmine clearing, and computerized/robotized surgery.