Home About us Contact
|
Home About us Contact | ||
|
|
||
Feedback Connections (feedback + connection)
Selected AbstractsIndividualized and time-variant model for the functional link between thermoregulation and sleep onsetJOURNAL OF SLEEP RESEARCH, Issue 2 2006STIJN QUANTEN Summary This study makes use of control system model identification techniques to examine the relationship between thermoregulation and sleep regulation. Specifically, data-based mechanistic (DBM) modelling is used to formulate and experimentally test the hypothesis, put forth by Gilbert et al. [Sleep Med. Rev.8 (2004) 81], that there exists a connection between distal heat loss and sleepiness. Six healthy sleepers each spent three nights and the following day in the sleep laboratory: an adaptation, a cognitive arousal and a neutral testing day. In the cognitive arousal condition, a visit of a television camera crew took place and subjects were asked to be interviewed. During each of the three 25-min driving simulator tasks per day, the distal-to-proximal gradient and the electroencephalogram are recorded. It is observed from these experimental data that there exists a feedback connection between thermoregulation and sleep. In addition to providing experimental evidence in support of the Gilbert et al. (2004) hypothesis, the authors propose that the nature of the feedback connection is determined by the nature of sleep/wake state (i.e. NREM sleep versus unwanted sleepiness in active subjects). Besides this, an individualized and time-variant model for the linkage between thermoregulation and sleep onset is presented. This compact model feeds on real-time data regarding distal heat loss and sleepiness and contains a physically meaningful parameter that delivers an individual- and time-depending quantification of a well known biological features in the field of thermoregulation: the thermoregulatory error signal Thypo(t),Tset(t). A validation of these physical/biological features emphasizes the reliability and power of DBM in describing individual differences related to the sleep process. [source] Memory effects description by neural networks with delayed feedback connectionsINTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, Issue 4 2004Petia D. Koprinkova For the purpose of dynamic systems modeling, it was proposed to include feedback connections or delay elements in the classical feed-forward neural network structure so that the present output of the neural network depends on its previous values. These delay elements can be connected to the hidden and/or output neurons of the main neural network. Each delay element gets a value of a state variable at a past time instant and keeps this value during a single sampling period. The groups of delay elements record the values of the state variables for a given time period in the past. Changing the number of the delay elements, which belongs to one group, a shorter or a longer time period in the past can be accounted for. Thus, the connection weights determine the influence of the past process states on the present state in a similar way as it is in the time delay kernel or cause-effect relation membership function (CER-MF) models. Specific feed-forward neural networks with time delay connections are used to solve the problem of neural network chemostat modeling as well as specific kinetic rates modeling. The weights of the feedback connections obtained during model training are discussed as the points of a time delay kernel or as the strength levels in a CER model (the points in the CER-MF). The corresponding changes in these weights with the changing time period in the past are shown. © 2004 Wiley Periodicals, Inc. [source] Dexterous manipulation of an object by means of multi-DOF robotic fingers with soft tipsJOURNAL OF FIELD ROBOTICS (FORMERLY JOURNAL OF ROBOTIC SYSTEMS), Issue 7 2002Pham Thuc Anh Nguyen This article analyzes the dynamics of motion of various setups of two multiple degree-of-freedom (DOF) fingers that have soft tips, in fine manipulation of an object, and shows performances of their motions via computer simulation. A mathematical model of these dynamics is described as a system of nonlinear differential equations expressing motion of the overall fingers-object system together with algebraic constraints due to tight area contacts between the finger-tips and surfaces of the object. First, problems of (1) dynamic, stable grasping and (2) regulation of the object rotational angle by means of a setup of dual two-DOF fingers, are treated. Second, the problem of regulating the position of the object mass center by means of a pair of two-DOF and three-DOF fingers is considered. Third, a set of dual three-DOF fingers is treated, in order to let it perform a sophisticated task, which is specified by a periodic pattern of the object posture and a constant internal force. In any case, there exist sensory-motor coordinations, which are described by analytic feedback connections from sensing to actions at finger joints. In the cases of setpoint control problems, convergences of motion to secure grasping together with the specified object rotational angle and/or the specified object mass center position, are proved theoretically. A constraint stabilization method (CSM) is used for solving numerically the differential algebraic equations to show performances of the proposed sensory-feedback schemes. © 2002 Wiley Periodicals, Inc. [source] Fault Diagnosis Based on the Fuzzy-Recurrent Neural NetworkASIAN JOURNAL OF CONTROL, Issue 2 2001Zhao Xiang ABSTRACT A fuzzy-recurrent neural network (FRNN) has been constructed by adding some feedback connections to a feedforward fuzzy neural network (FNN). The FRNN expands the modeling ability of a FNN in order to deal with temporal problems. A basic concept of the FRNN is first to use process or expert knowledge, including appropriate fuzzy logic rules and membership functions, to construct an initial structure and to then use parameter-learning algorithms to fine-tune the membership functions and other parameters. Its recurrent property makes it suitable for dealing with temporal problems, such as on-line fault diagnosis. In addition, it also provides human-understandable meaning to the normal feedforward multilayer neural network, in which the internal units are always opaque to users. In a word, the trained FRNN has good interpreting ability and one-step-ahead predicting ability. To demonstrate the performance of the FRNN in diagnosis, a comparison is made with a conventional feedforward network. The efficiency of the FRNN is verified by the results. [source] | ||||||||||