Sven Nomm

Recognition of the Surgeon's Motions During Endoscopic Operation by Statistics Based Algorithm and Neural Networks Based ANARX Models

Abstract The problem of recognition and short time prediction of the surgeons hand motions during surgical endoscopic operation are approached in the present contribution using neural network based nonlinear modeling techniques and statistics based segmentation of the operating room. It is shown that proposed technique provide precise recognition of surgeons motions.

Monitoring of the Human Motor Functions Rehabilitation by Neural Networks Based System with Kinect Sensor

Abstract Application of the Kinect sensor for the automatic monitoring and supervision of the human motor functions rehabilitation is considered in this paper. Information about position of patient limbs provided by Kinect sensor is analyzed by the neural networks based system which determines if therapeutic exercise is performed in correct or incorrect way.

Neural Networks Based ANARX Structure for Identification and Model Based Control

This article is devoted to the training and application of neural networks based additive nonlinear autoregressive exogenous (NN-based ANARX) model. Training of NN-based ANARX model with MATLAB is discussed in detail and illustrated by examples. Dynamic state feedback linearization control algorithm is then applied for control of unknown nonlinear system

Linear Input-Output Equivalence and Row Reducedness of Discrete-Time Nonlinear Systems

The problem of linear input-output (i/o) equivalence of mero morphic nonlinear control systems, described by implicit higher order difference equations, is studied. It is proved that any system is linearly i/o equivalent to a row-reduced form. The constructive algorithm is given for finding the required transformation. The latter amounts to 1) multiply the set of i/o equations ψ = 0 from left by a unimodular matrix A(δ), whose entries are non-commutative polynomials in the forward-shift operator δ, and 2) define certain multiplicative subset of the difference ring of analytic functions which introduces some inequations that should be satisfied.

A novel taylor series based approach for control computation in NN-ANARX structure based control of nonlinear systems

This paper presents an alternative approach for control computation in a closed loop of discrete-time nonlinear system and NN-ANARX based dynamic output feedback. Proposed technique is based on an application of Taylor series expansion for computation of control directly from neural network based model. Two modifications of the algorithm are proposed for both single-input single-output and multi-input multi-output nonlinear systems. The effectiveness of the proposed approach is demonstrated on numerical examples.

Adaptive Output Feedback Linearization for a Class of NN-based ANARX Models

Present paper is devoted to the design of an adaptive output feedback controller for nonlinear system modelled by neural networks based Additive Nonlinear Autoregressive Exogenous structure. Off-line and on-line parameter identification of the neural networks based Additive Nonlinear Autoregressive Exogenous model using standard training algorithms are discussed in detail and illustrated by numerical simulations. The main contribution of this paper is in combining neural networks based adaptation with dynamic output feedback linearization technique.

Application of self organizing Kohonen map to detection of surgeon motions during endoscopic surgery

Segmentation of the surgeonpsilas hand movements during the surgery into more primitive parts and recognition of those parts using Kohonen map is discussed in present paper. Main advantages of the proposed approach are that it allows to take into account dynamical characteristics of the hand movements and exclude probability of human error in building etalon segmentation. Ability to recognize current action of the surgeon has a crucial importance in developing a robot able to assist surgeon during the endoscopic surgical operation. One of the possible ways is to predefine a set of possible surgeonpsilas actions and provide a recognition algorithm explored in the framework of present contribution.

Irreducibility Conditions for Continuous-time Multi-input Multi-output Nonlinear Systems

The purpose of this paper is to present necessary and sufficient condition for irreducibility of continuous-time nonlinear multi-input multi-output system. The condition is presented in terms of the greatest common left divisor of two polynomial matrices related to the input-output equations of the system. The basic difference is that unlike the linear case the elements of the polynomial matrices belong to a non-commutative polynomial ring. This condition provides a basis for finding the equivalent minimal irreducible representation of the I/O equations which is a suitable starting point for constructing an observable and accessible state space realization

Comparison of neural networks-based ANARX and NARX models by application of correlation tests

A correlation-test-based validation procedure is applied in this study to compare neural networks based nonlinear autoregressive exogenous model class to its subclass of additive nonlinear autoregressive exogenous models.

Dynamic output feedback linearization based adaptive control of nonlinear MIMO systems

This paper discusses application of dynamic output feedback linearization algorithm for adaptive control of nonlinear MIMO systems. Neural Network based Simplified Additive Nonlinear AutoRegressive exogenous (NN-SANARX) structure is used for identification of nonlinear MIMO systems. This structure imposes a restriction on model adaptation. The model is divided into adaptable and nonadaptable parts. After that history-stack adaptation with dynamic output feedback linearization is used for adaptive control of nonlinear MIMO systems. The effectiveness of the adaptive control technique proposed in the paper is demonstrated on numerical example.