Stefano Di Gennaro

Sensorless High Order Sliding Mode Control of Induction Motors With Core Loss

In this paper, a sensorless control scheme is presented for induction motors with core loss. First, a controller is designed using a high order sliding mode twisting algorithm, to track a desired rotor velocity signal and an optimal rotor flux modulus, minimizing the power loss in copper and core. Then, a super-twisting (ST) sliding mode observer for stator current is designed and the rotor flux is calculated, by means of the equivalent control method. Two methods for the rotor velocity estimation are then proposed. The first consists of a further super-twisting sliding mode observer for rotor fluxes, with the purpose of retrieving the back-electromotive force components by means of the equivalent control method. These components are functions of the rotor velocity which, hence, can be easily determined. The second method is based on a generalization of the phase-locked loop methodology. Finally, a simple Luenberger observer is designed, filtering the rotor velocity estimate and giving also an estimate of the load torque. The performance of the motor is verified by means of numeric simulations and experimental tests, where good tracking results are obtained.

Detection of epileptiform activity in EEG signals based on time-frequency and non-linear analysis

We present a new technique for detection of epileptiform activity in EEG signals. After preprocessing of EEG signals we extract representative features in time, frequency and time-frequency domain as well as using non-linear analysis. The features are extracted in a few frequency sub-bands of clinical interest since these sub-bands showed much better discriminatory characteristics compared with the whole frequency band. Then we optimally reduce the dimension of feature space to two using scatter matrices. A decision about the presence of epileptiform activity in EEG signals is made by quadratic classifiers designed in the reduced two-dimensional feature space. The accuracy of the technique was tested on three sets of electroencephalographic (EEG) signals recorded at the University Hospital Bonn: surface EEG signals from healthy volunteers, intracranial EEG signals from the epilepsy patients during the seizure free interval from within the seizure focus and intracranial EEG signals of epileptic seizures also from within the seizure focus. An overall detection accuracy of 98.7% was achieved.

Classification of EEG signals for detection of epileptic seizures based on wavelets and statistical pattern recognition

The electroencephalogram (EEG) signal is very important in the diagnosis of epilepsy. Long-term EEG recordings of an epileptic patient contain a huge amount of EEG data. The detection of epileptic activity is, therefore, a very demanding process that requires a detailed analysis of the entire length of the EEG data, usually performed by an expert. This paper describes an automated classification of EEG signals for the detection of epileptic seizures using wavelet transform and statistical pattern recognition. The decision making process is comprised of three main stages: (a) feature extraction based on wavelet transform, (b) feature space dimension reduction using scatter matrices and (c) classification by quadratic classifiers. The proposed methodology was applied on EEG data sets that belong to three subject groups: (a) healthy subjects, (b) epileptic subjects during a seizure-free interval and (c) epileptic subjects during a seizure. An overall classification accuracy of 99% was achieved. The results confirmed that the proposed algorithm has a potential in the classification of EEG signals and detection of epileptic seizures, and could thus further improve the diagnosis of epilepsy.

Adaptive integrated vehicle control using active front steering and rear torque vectoring

This work studies the combination of Active Front Steering (AFS) with Rear Torque Vectoring (RTV) actuators in an integrated controller to guarantee vehicle stability. Adaptive feedback technique has been used to design the controller. The feedback linearisation is applied to cancel the nonlinearities in the input?output dynamics of the vehicle. Parameter adaptation then is used to robustify the exact cancellation of the nonlinear terms. The results show tracking and stabilisation capabilities when important parameters, like tyre stiffness and tyre characteristics, are affected by estimation errors.

Theory of Optimal Control Using Bisimulations

We consider the synthesis of optimal controls for continuous feedback systems by recasting the problem to a hybrid optimal control problem: to synthesize optimal enabling conditions for switching between locations in which the control is constant. An algorithmic solution is obtained by translating the hybrid automaton to a finite automaton using a bisimulation and formulating a dynamic programming problem with extra conditions to ensure non-Zenoness of trajectories. We show that the discrete value function converges to the viscosity solution of the Hamilton-Jacobi-Bellman equation as a discretization parameter tends to zero.

Critical Observability of a Class of Hybrid Systems and Application to Air Traffic Management

We present a novel observability notion for switching systems that model safety–critical systems, where a set of states – called critical states – must be detected within a prescribed delay since they correspond to hazards that may yield catastrophic events. Some sufficient and some necessary conditions for critical observability are derived. An observer is proposed for reconstructing the hybrid state evolution of the switching system whenever a critical state is reached. We apply our results to the runway crossing control problem, i.e., the control of aircraft that cross landing or take–off runways. In the hybrid model of the system, five agents are present; four are humans, each modeled as hybrid systems, subject to situation awareness errors.

Efficient Solution of Optimal Control Problems Using Hybrid Systems

We consider the synthesis of optimal controls for continuous feedback systems by recasting the problem to a hybrid optimal control problem: synthesize optimal enabling conditions for switching between locations in which the control is constant. An algorithmic solution is obtained by translating the hybrid automaton to a finite automaton using a bisimulation and formulating a dynamic programming problem with extra conditions to ensure non-Zenoness of trajectories. We show that the discrete value function converges to the viscosity solution of the Hamilton--Jacobi--Bellman equation as a discretization parameter tends to zero.

Optimal Control Using Bisimulations: Implementation

We consider the synthesis of optimal controls for continuous feedback systems by recasting the problem to a hybrid optimal control problem which is to synthesize optimal enabling conditions for switching between locations in which the control is constant. We provide a single-pass algorithm to solve the dynamic programming problem that arises, with added constraints to ensure non-Zeno trajectories.

Observability of Switched Linear Systems: A geometric approach

In this work, the observability of continuous-time Switched Linear Systems (SLS) is studied using a geometric approach. The approach herein presented allows representing in a common framework reported results that are based on inferring the evolving linear system from the knowledge of the continuous inputs and outputs. Also, necessary and sufficient conditions for the observability of SLS with unknown inputs are presented, which to the best of our knowledge has not been characterized so far for this class of dynamic systems.

Verification of Hybrid Automata Diagnosability With Measurement Uncertainty

The problem of system diagnosability verification is concerned with whether a fault in the system operation can be diagnosed by using the system model and observations of the system output. In this paper, we investigate the (δd, δm)-diagnosability of hybrid automata, which characterizes the maximum delay for diagnosing faults since their first occurrence, given the measurement uncertainty in observing the system output. We present a methodology that analyzes the (δd, δm)-diagnosability of hybrid automata. Due to the complex dynamics, the hybrid system diagnosability is often difficult to analyze directly. We thus propose an approach of constructing an abstraction using the trajectories of the original system. Their (δd, δm)-diagnosability properties are proved to be quantitatively related to each other. The abstraction has only finitely many trajectories that extend to the end of the time horizon of interest, so its diagnosability can be easily calculated, and then used to derive the diagnosability of the original system. We illustrate this procedure with an example.