Julian Stoev

  • Citations Per Year
Learn More
The paper is extending output feedback nonlinear control and backstepping approaches to a class of systems approximately di2eomorphic to output feedback systems that include unknown functions. The unknown functions are addressed via online function approximation, which results in two types of uncertainty. Parametric uncertainty due to online function(More)
This paper presents an overview of model-based (Iterative Learning Control, Model Predictive Control and Iterative Optimization) and non-model-based (Genetic-based Machine Learning and Reinforcement Learning) learning strategies for the control of wet clutches. Based on theoretical considerations and a validation on an experimental test bench containing wet(More)
A common approach when applying reinforcement learning to address control problems is that of first learning a policy based on an approximated model of the plant, whose behavior can be quickly and safely explored in simulation; and then implementing the obtained policy to control the actual plant. Here we follow this approach to learn to engage a(More)
We present a Mechatronics design approach and related technologies for a badminton playing robot, as a first stage of a multi-year project. The robot is using non-modified shuttles and rackets, which are detected and localized using purely visual information. The robot subsystems are presented: mechanical design, visual detection of the shuttle, shuttle(More)
Modeling of hydraulic clutch transmissions are not straightforward because of their hard nonlinearities. Therefore a model-free optimizer is used to tune the input parameters for desired operation, thereby generating optimal reference trajectory. However these machines suffer from time varying dynamics, hence a feedback predictive control is used to track(More)
This paper discusses energy optimal point-to-point motion control for linear time-invariant (LTI) systems using energyoptimal Model Predictive Control (EOMPC). The developed EOMPC, which is based on time-optimal MPC, aims at performing energy-optimal point-to-point motions within a required motion time. Energy optimality is achieved by setting the object(More)
A model order reduction algorithm is derived, which is suitable for linear systems with a large number of inputs and outputs. The full system matrices may be non-symmetric, singular and do not require other properties like positive definiteness. Such systems typically arise from finite element discretization of partial differential equations describing(More)
It is well known that universal approximators can be used for adaptive control and estimation. In this paper, the problem of adaptive state observation of a large class of nonlinear uncertain systems is considered and it is shown that splines have some special properties, which can lead to simplified observer structure. In particular, the observer filter(More)