Introduction To Physiological Control Systems PDF

1 INTRODUCTION TO PHYSIOLOGICAL CONTROL A control system is generally defined as a collection of interconnected components used to achieve a desired response in spite of external disturbances. Control and Regula...

1 INTRODUCTION TO PHYSIOLOGICAL CONTROL A control system is generally defined as a collection of interconnected components used to achieve a desired response in spite of external disturbances. Control and Regulation are two faces of the same medal. Control emphasizes the action of bringing the system to the wanted status point, thus obtaining the desired output. Regulation stresses the modification of the controlled system properties. E.g. its stability or the rapidity in compensating disturbances. There are two ways in which a control system can be made to operate: 1. Open-loop control system Open loop regulation may perform complex behaviours but cannot compensate disturbances, that are a-priori unknown. E.g. a programmable thermostat can correctly predict day and night time thus adapting the heating system, but can’t compensate the changes due to good or bad weather. In general, open-loop control can be said to predictive. It is important to acknowledge that the complex actions we can do are mainly programmed in our neural circuits, in some part innate, but mostly acquired by experience and training. However, without feedbacks they could lead us to wrong actions. E.g., you may have experienced the wrong gesture of rising too much a suitcase you supposed but which was conversely light! Only the sensory feedback permits to correct that! (see next). A particular case of open-loop controls (mainly in artificial controls) is that of servomechanisms, which amplify a driving force, based on the open-loop forecast that a force multiplied by a fixed coefficient is necessary. E.g. the power steering (it. servosterzo) in a car.  2 2. Closed-loop control system Closed-loop is a core element of automatic control and regulation in both artificial and natural systems. It is based on the concept of feedback. FEEDBACK: “a method for controlling a system by reinserting into it the results of its past performance”. This seminal statement is due to Norbert Wiener (1940s) working at the MIT, Boston, starting the entire discipline of Automatic Control (alias Automation, alias Automatica). Disturbance x Controller Output y Input r + Error e action u Plant Controller (controlled system) - Feedback signal z Feedback sensor Example: thermostat (controller) - regulation of room temperature - temperature sensor (feedback) - error compared to a set point - control of fan-heater (plant, it. impianto). CYBERNETICS: control theory applied to biological and artificial systems (Greek kubernetes, Latin gubernator = steersman, it. timoniere). This term was introduced by Norbert Wiener as a paradigm of feedback control: the controlled plant is the ship; the controlled variable is the actual ship direction; disturbances are the changes in the wind, sea currents and waves, etc.; reference is made by an aim at the horizon, or a star, or the compass; the feedback measure is acquired by the steersman and compared with the reference, thus evaluating an error; the steersman is the controller inserted in the loop; the control variable is the steer position. Although this term is now obsolete, it is culturally important to see how Wiener was considering both artificial and natural control systems since the very beginning of Automatic Control.  3 Integration of Open and Closed-loop controls A first sign of complexity in our regulation mechanisms is that practically all controls do have an open loop basis on top of which feedbacks are acting. A nice example could be that of sport jumpers trained by repeating again and again the same gestures which permit them to go over the obstacle. This is the way by which their motor system memorizes a very difficult open loop control action. However, the same athletes are also able to compensate unexpected changes (e.g., the wind), which is done by feedbacks. In this course, emphasis will be given to closed-loop, but this integration should be never forgotten. Also the autonomic cardiovascular regulation, though based on closed- loop feedbacks, includes relevant predictive mechanisms. E.g., the respiratory, cardiovascular, muscular system of an athlete ahead of the race is start are predictively activated (sympathetic arousal, see next). This happens also when we stand-up from sitting or supine position. In fact, passively raising a person by a tilting bed from supine to standing lacks this predictive action. All is based on feedbacks. And if such regulation is poorly responding the person may faint (it. svenire) having a vasovagal syncope. The physiological control systems have many of these complex integrations, to be illustrated at the end of this chapter by examples (see below).  4 In Physiology: HOMEOSTASIS - “It is the fixity of the ‘milieu interieur’ (It. ambiente interno) which is the condition of free and independent life.” (Claude Bernard, 19th century). The intuition that the life of our cells and tissues needs stable conditions was the turning point to acknowledge that this was obtained by regulation mechanisms and control feedbacks. Hence, the concept of feedback steadily entered physiology and was developed in the 20th century by scholars as Guyton and Mountcastle. Homeostasis in the milieu interieur does not mean absence of dynamics in a living organism. “stasis” ≠ “statics” The goal of homeostasis requires to get food, water, oxygen, protection from external temperature, etc., etc. All of this is achieved by coordinated actions, each of them contains complex regulatory mechanisms, both in open-loop and closed-loop. The complex regulatory mechanisms of the whole body actively counteract the external disturbances and pursue vital goals necessary to life (searching food, air, acceptable climatic conditions, etc.). However, all these activities are ultimately directed to keep the stasis of the internal cellular ambient. We can say that homeostasis is the core reference to all the intrinsic set points of our control mechanisms.  5 REFLEX VS FEEDBACK: two simple reflex examples and additional considerations 1 – The Muscle Stretch Reflex Schematic Illustration of the muscle stretch reflex Tap-induced Stretch, x (Disturbance) Change in Efferent Reflex Neural Frequency, u Center Thigh Extensor (Spinal Cord) Muscle (Controller action) Change in Afferent Change in Muscle Neural Frequency, z Spindle Length, y Muscle Spindle (Feedback signal) (System Output) Ed back se Block diagram representation of the muscle stretch reflex (it. “riflesso patellare”) (thigh extensor = estensore della coscia; muscle spindle = fuso muscolare)  6 2 – The Baroreflex The baroreflex is one of the main regulation mechanisms in the cardiovascular (CV) system (together with the chemoreflex, the cardio- pulmonary reflexes, muscular activation reflexes, etc.) Literally, it means: reaction (reflex) to pressure (baro). More completely, reflex reaction to arterial pressure changes. It starts from arterial stretch receptors in high-pressure areas (carotid sinuses, aortic arch, mainly). To be not confounded with other reaction to pressure changes starting from the cardiopulmonary and other low pressure areas. Block diagram of Baroreflex directed to the vessels and to the heart. SAP = Systolic Arterial Pressure, RR = RR Interval (Heart Period duration), PR = Peripheral Resistance, CNS =Central Nervous System, CNS = Central Nervous System (mainly cardiovascular centres in the medulla, it. “bulbo”) In the highly simplified scheme of this figure we note that: the plant (action mechanisms) includes at least two branches: one directed to the vessels, adapting pressure by vasoconstriction; the second directed to the heart (specifically the sinus node), adapting the cardiac period (hence its inverse, the heart rate); a measure is drawn from the pressure produced by heart and vessels; a feedback is sent by baroreceptive fibres to the CNS; a (virtual) comparison with an ideal set-point pressure is performed; a control is output through two pathways: (+) sympathetic, (-) parasympathetic.  7 Note about the “REFLEX” concept In the physiological language, Reflex is often used as synonym of feedback within physiological control systems. Actually, reflex suggests a response to some change (provocation in clinical/physiological studies; transient in automation and signal processing). This is not always the case for feedbacks, which in many control systems and instances work against ongoing (stationary) disturbances and output oscillations. The reason why physiologists prefer the term reflex is both historical and experimental: feedbacks are easily observed as reactions to some provocative stimulus; this can be more easily reproduced in experiments and unknown reflexes can be searched for or known ones can be assessed more easily by observing the transient response (reflex) to a controlled stimulus. Indeed, the feedback loop is similar in both transient and ongoing responses, and in normal conditions physiological control mechanisms are working against the latter case. A main difficulty in studying physiological control is in putting together data relevant to transient responses and data concerning the ongoing activity. In the following, we will come through the baroreflex several time and in several ways: considering transients over short times, considering the frequency response to periodic inputs (e.g. respiration cycle), considering ongoing activity (closed loop identification).  8 PHYSIOLOGICAL CONTROL SYSTEMS (VS ARTIFICIAL ONES) 1) Physiological VERSATILITY (vs fixed tasks): integration of various open loop and closed loop regulation strategies. 2) Unknown components à SYSTEM IDENTIFICATION à laws derived from data (black-box) vs. internal models based on known artificial components (white-box). Even when we know some physiological law based on physical considerations we don’t know the value of the parameters for that specific subject in that specific condition. This is called a “gray-box”, the parameters of which must be identified by fitting specific data. 3) Extensive CROSS-COUPLING (it. interazioni): Example: contributions of interrelated systems to the muscle stretch reflex More examples: arterial pressure control - low pressure and high pressure areas - blood volume control - control of flow in organs and limbs - thermo-regulation - renal activity - intracranial pressure … 4) ADAPTIVE control mechanisms and HIERARCHICAL structure:  9 Example: adaptive characteristics of the muscle stretch reflex 5) EMBEDDED (it. “intrinseco”) negative feedback and set point: Note: we will go into this counter-intuitive though central concept in the “Equilibrium chapter” and provide examples. Definition: creation of feedback working point not through a comparison of measure and an external set-point but by satisfying feedforward and feedback non-linear static characteristics. By linearizing around the working point, this can be represented as a virtual “set-point”, however an embedded not an external one! Adaptation of the working point is obtained by adapting the control loop non-linear features (see examples in the Equilibrium chapter), not by moving an external set-point (who should move it?!). 6) Generally, NON-LINEAR features of both control and plant.