A linear, simple robot is one of the easiest ones to handle and control. When the controller has a complete understanding of goals as well as variables, it has complete capability to direct the robot with rules that need to be followed. So, it knows that if user is pressing button 1, the robot needs to pick up an object and leave the same when some other button is pressed. The difficulty lies with the non-linear, complicated robots. The rules change more frequently as compared to the variables and goals none of which are clear to anyone.
According to Zhijun Fu, a researcher from Zhejiang University’s Department of mechanical engineering, China, “The knowledge of system dynamics is completely unknown and system states are not available… therefore, it is desirable to design a novel control scheme that does not need the exact knowledge of system dynamics but only the input and output data measured during the operation of the system.” The researchers needed to determine the states of systems to figure out a way to control these. To accomplish this, they implied a neural network that helped observe the framework at multiple time scales and update the information as they progressed their study.
Fu further adds, “We cannot apply existing actor-based methods to unknown nonlinear systems directly. An actor must be told what to do, while the observer watches the system to learn the requirements for optimal control. The proposed method may be used [in] industrial systems with ‘slow’ and ‘fast’ dynamics, due to the presence of some… parameters, such as small time constants.” Variables of these type can really enhance the cost of complete system with respect to resources as well as energy. The observer-based method considers every possible parameter and makes the adjustments according to that. The method also addresses a very common problem associated with system control – the overwhelming behavior of actuators. In automated machines, the actuators often get saturated and start malfunctioning or stop working completely. When you account for input constraints a control system allows actuators to keep oversaturation at bay.
Filed Under: News