Abstract
The appeal of flexible robots that are lightweight and can operate safely near humans is on the rise. However, because of the flexibility aspect, vibration throughout the robot’s motion becomes a major issue. The purpose of this Master’s thesis is to understand how to control and suppress vibration on a flexible robot. There are various control techniques to control a flexible robot and suppress vibration, however, the active open-loop method is the primary focus of this research. A physical 1 degree of freedom (DOF) flexible link robot manipulator and Simulink models of this physical system were developed to run simulations and experiments to observe and test three different input commands. The three input commands are a standard step input, a 6th-order polynomial S-curve input, and the utilization of a Gopinath-style velocity state observer (with a step input). The robot manipulator was designed to carry a transparent container with water, whose initial value is 0.085 kg, intending to transport it from 0 to 90 degrees without spilling. The results show that the standard step input experienced the most vibration and lost an average of 0.0296 kg. Both the 6th-order polynomial S-curve input and the state observer method did not lose any water during the motion. However, between these two, the 6th-order polynomial S-curve input experience the least vibration and was able to reach the desired location faster than the state observer method. In conclusion, although the 6th-order polynomial S-curve input performed better than the state observer method, the 6th-order polynomial S-curve input was specifically tailored to the given parameters of the robot manipulator. Future work is needed in order to test out the range of parameter values that the state observer method can successfully operate under.