Browsing by Author "Kilicaslan, Sinan"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Article Citation - WoS: 2Citation - Scopus: 2Hybrid Force and Motion Control of a Three-Dimensional Flexible Robot Considering Measurement Noises(Mdpi, 2022) Kilicaslan, Sinan; Ozgoren, Mustafa Kemal; Ider, Sitki KemalThis work addresses the end-effector trajectory-tracking force and motion control of a three-dimensional three-link robot considering measurement noises. The last two links of the manipulator are considered as structurally flexible. An absolute coordinate approach is used while obtaining the dynamic equations to avoid complex dynamic equations. In this approach, each link is modeled as if there is no connection between the links. Then, joint connections are expressed as constraint equations. After that, these constraint equations are used in dynamic equations to decrease the number of equations. Then, the resulting dynamic equations are transformed into a form which is suitable for controller design. Furthermore, the dynamic equations are divided as pseudostatic equilibrium and deviation equations. The control torques resulting from the pseudostatic equilibrium and the elastic deflections are obtained easily as the solution of algebraic equations. On the other hand, the control torques corresponding to the deviations are obtained without any linearization. Encoders, strain gauges, position sensors and force and moment sensors are required for measurements. Low pass filters are considered for the sensors. For the crossover frequencies of the sensors, low and high values are chosen to observe the filtering effect on the robot output.Article Citation - WoS: 1Citation - Scopus: 1Motion Control of a Spatial Elastic Manipulator in the Presence of Measurement Noises(Springer Heidelberg, 2021) Kilicaslan, Sinan; Ider, S. Kemal; Ozgoren, M. KemalThis paper presents a method for the end effector motion control of a spatial three-link robot having elastic second and third links including measurement noises. In the derivation of equations of motion, not to face with complex equations of motion, each link is modeled as though the links are not connected and the restrictions on the links due to connecting them by joints are written as constraint equations. After that the Lagrange multipliers are eliminated and the constraint equations at the acceleration level are substituted into the equations of motion to reduce the number of equations. To handle the non-minimum phase property, the equations of motion of the elastic manipulator are divided as the equations corresponding to a pseudostatic equilibrium and the equations of the deviations from them. Definition of the pseudostatic equilibrium used in this study can be given as a hypothetical state in which the end effector velocity and the end effector acceleration possess their reference values while the elastic deflections are instantly constant. The advantages of this control method are that the elastic deflections and the control inputs required for the pseudostatic equilibrium are obtained by an algebraic method and the feedback stabilization control inputs for the deviation equations are determined without linearizing the dynamic equations. The required measurements are obtained from the strain gauges on the links, the encoders placed on the joints and the position sensors attached to the end effector. For each sensor, a low pass filter is used. Simulations are made with low and high values of crossover frequencies to show the positive and negative effects of filtering on the responses of the system.

