| Summary: | In this thesis the problem of trajectory planning for automatic machines is addressed, considering in particular the problem of vibration suppression and perfect tracking. Two different trajectory generators based on dynamic filters are developed and implemented. These novel trajectory planners are designed respectively for residual vibrations suppression and perfect tracking of periodic trajectories. Both solutions are very effective and easy to implement, exploiting the realization of dynamic filters by means of FIR filters.
In the first part of the thesis the problem of residual vibrations in motion control of robots is addressed. In a review fashion the most widespread feed-forward techniques for residual vibration suppression are analyzed in detail and implemented in simulation. Moreover all the analysis are performed with a control system perspective in order to give a unified point of view allowing to compare all the solutions despite their differences.
In Chapter 5 a novel trajectory generator based on Exponential Filters is presented, analyzed and compared with the most commonly used feed-forward techniques for vibration suppression, proving comparable performances with the state of the art. The analysis and comparison procedure is performed both in simulation and in experimental activities. This new method is developed first for simple SISO LTI systems and then extended to MIMO systems.
In Chapter 6 a new repetitive control scheme based on B-Spline Trajectory Generator that exploits dynamic filters is presented. The novel scheme integrates the trajectory generator and the repetitive controller in a single discrete time feedback loop achieving perfect tracking for periodic motions.
In appendix, the development of a 6-axis Force/Torque sensor for underwater activities based on optoelectronic components is reported. The description covers the entire process from the concept to the development of a simulation model and finally to the prototype realization, along with an extensive experimental activity.
|
|---|
