INTRODUCTION TO THE CONTROL PROBLEM Preview In this chapter, our aim is to seek answers to questions like, why is control important? significance of feedback? In addition, our objective is to learn the terms commonly used in ¢ system engineering and the basic features and configurations of control systems. A rich ¥ practical problems are placed as examples which serve the dual purpose of (a) showing t of control theory, and (b) defining the basic structures of control systems. i CONTROL SYSTEMS: TERMINOLOGY . __ AND BASIC STRUCTURE and advancement of modern civilization and technology. Practically, every activities is affected by some type of control system. A home-heating system, a machine, a refrigerator, an air conditioner, and an automobile are all examples 0 trol systems are indispensable in modern industrial processes. We find control the industry, such as quality control of manufactured products, automatic control, space technology and weapon systems, transportation systems, many others. Even such problems as inventory control, and socio-eco approached from the theory of feedback control. In this book, we will d ing systems that are governed by the laws of physics and are therefore calle In control parlance, the system to be controlled is given various names: system being perhaps the most common. In the so-called process industri power, fuel, etc.), one repeatedly encounters. the need to control temy vessels, pressure, humidity, chemical composition, and the like, Such sidered as process control applications. Historically, the wide practical place in the process area. Most of the basic concepts were developed an tion by the intuitive and experimentally oriented engineering metho ‘Around the time of the Second World War, the technical ne matic airplane pilots, gun-positioning systems, radar antenna cont more scientific approaches in the control engineering field. A. the theory of servomechanism'—aimed mainly at applications Wh cal motions (position, velocity, or acceleration), was developed design methods of the process-control area, and the servome The word servomechanism originated from the words *e-n systems: Principles and Design reas is also converging. The word ‘process’ is now in use for all types respective of the area of application, the word ‘servomechanism’ (0 ‘system wherein the controlled attribute of the system, ‘The terminology of the two a of controlled systems. Also, i servo system) is used for a command following, required to follow a given command. Figure 1.1 shows the input-output configuration of.a process (or plant). Process outputs are the Disturbaiion epee response variables which we require to behave in some specified fashion, Process inputs are flows of energy and/or material that cause the process to react or respond, The inputs are classified into manipulated inputs (subject to our control) and disturbance inputs (undesirable and unayoidable effects beyond our control, generated from outside the process environment, and from within). The presence of the disturbance is one of using the control. Judicious management of manipulated variables so as fo ¢o disturbances is the primary role of the controller. Figure 1,2 shows the input-o a control system. i Command 1 Manipulated input | bl SRE HU! | Controller F222 | Process 1 When the desired value of the controlled output is more or reject disturbance effects, the control system is sometimes a regulator becomes a constant and is called a setpoint, which co controlled output. The set-point may, however, be changed in tit and the need for the control of manipulated variables arises fron changes and the disturbance rejection, The control problem ist In the follow-up or tracking system, the controlled out varying command input. For such systems, the need for the the requirements of command following as well as disturban In the configuration of Fig. 1.3, the controller receives * controlled output and uses this information as a means of ec trast to this configuration, the configuration of Fi i Command inputDisturbance input Command input Manipulated variable Controlled output Feedback signal Measurement Fig. 1.4 Input-output configuration of a closed-loop control system difference as a means of control of manipulated variables. The configuration of Fig. open-loop control system while that of Fig. 1.4 is that of a closed-loop control systet control system. To make these definitions more concrete, let us consider some familiar examples of Example 1.1 Automobile driving system The system to be controlled has two inputs (steering and acceleration/brakin, puts (heading and speed). The two command inputs are the direction of the with trafic signals. A block diagram of this two-input, two-output system is sh Direction ‘Wind + traffic of highway is c Steering wheel position Actuator Control on (Hands) Actuator 7 GA 5 ie (Brain) (Foot) Speed Error limits detector Systems with more than one controlled output and command i 1, multi-oufput (MIMO) systems, On the other hand, tn sii gle output ig controlled by @ single input. In multivariable sya ® particular output also affeets the onet conte a his ; mis 5 The automobile-driving syste iunen tems for the purpose of. ‘design because the interaction is ne; inputve celerator control affects the speed and a 4 a pire decreases the side forces at the A aN ‘with locked wheels the directional control is completely lost, If we Consider this voetigile eo eaniettt Rem Vehicle-heading. control system as a decoupled Sigs, inveraction with he ee peed control system, Be The command inputs for an automobile on the toad cannot be constant depend on the traffic and road conditions and vary in an uncontrolled manner, Th to the system are derived by the driver from the actual road and traffic ¢ ic tem will therefore be a component in the overall control system of th in Fig. 1.5. The desired course (heading) and speed are compared with a course and speed in order to generate a measurement of the error. The hun manipulated variables (steering wheel position, accelerator/brake Position) which reduces the absolute error. not the heading. However tyre-road interface for d Example1.2 A heating system To this point, we have discussed in very general terms about the purpose of and closed-loop control strategies. To further understand these concepts, heating system fora chamber, used for testing manufactured produets for th system, the chamber temperature, 6(°C) is the Tesponse variable of interest manipulated variable is the input Power (kW) to heat the chamber. Inintroauuuir wy ane era ——_w i t temperature ‘Temperature Environment Ips a setpoint ea Chamber 0 Potetiometer| al Power ea Lf Stee ya Fee er Input power (manipulated variable) (b) Fig. 1.6 Open-loop temperature control system temperature, 8,, which is the command input, is set on a calibrated dial. This positions wiper that applies a voltage signal, e, (volts) to the power amplifier, which supplies # the heater element inside the ehamber. The chamber temperature is thus, controlled b therefore some heat is lo&t Airc ti the walls. The amount of heat flow through the temperature differential that exists across the walls and the thermal properties. re 1.6b shows the functional block diagram of the open-loop temperature The dial is calibrated when the environment temperature has certain value: to the power required to regulate the chamber temperature as per command loss to the environment. If the environment temperature remains constant time of calibration, then the open-loop temperature control scheme of Fig. J performance. However, the environment temperature does fluctuate, and from the desired value by a large error, and hence, precise control ¥ control scheme of Fig. 1.6 is used when performance requirements are loop scheme, there is no automatic adjustment of any errors in outp We have to live with the consequences or redesign/recalibrate The way to realize precise control is to replace the open-loop closed-loop control scheme, that can act on the information receiy wrong. By measuring the process output (chamber temperature), we are closing the loop. Figure 1.7 gives one such feedback control scheme. For temp sensor which has, in addition to the basic device such asa the1 \eak signals from thermocouple. The output volts e,, of the reference voltage signal e,; to make it compatible with e, for eo ‘or to generate the difference between the reference signall/@,a SereneEnvironment (@y) Chamber (@) Voltage amplifier Thermocouple Heater Power Voltage amplifier amplifier T rence voltage from the comparator is amplified and applied current to drive the heater element inside the chamber. ‘at causes the temperature @ as well as the feedback ‘ven command input), and the difference signal (e, =€,) to As the difference signal approaches the zero value, supply of heat to the chamber is also cut off. As a resi o the reference signal e, (ie., 0= 8,). ‘mal equilibrium within the chamber just establi joss to the environment, It is a matter of time, x the temperature within the chamber to droj s restored, thus causing the error signal (e, = ¢,) t0 Thus, through the use of feedback, the system monitol show up the effects of the disturbances) and takes cos prov a automatic control of the chamber temperatui desired value, or atleast close to the desired value, Figural closed-loop temperature control system, — Comparison of Figs 1.5 However, the two systems di and 1.8 gives many common) ffer clearly in one aspect, In Fi outputs (actual course, and speed of travel) for comp kReference Error Control Manipulated voltage, e, voltage signal power Function for ; calculation Voltage Power Ore amplifer | "| amplifer Feedback Signal, e Fig. 1.8 Functional block diagram of the system of Fig. 1.7 Temperature sensor and speed of travel), and the control actions (rotation of steering wheel, and accel take place through the human operator. In Fig. 1.8, the human operator has been matic error-detecting and correcting devices; both feedback of the controlled 0 temperature), and the correcting action (control of the input power) take place ¥ a human operator. Such systems, therefore, are automatic closed-loop control The familiar examples of feedback control cited earlier, fit into the basic str To understand Fig, 1.9, we begin with the command signal y,—the des variable y. This signal has the same units as the controlled variable (a physic Elements ‘A; Reference function elements yy : D_ = Control function elements r Gy: Actuator elements a Gp ; Controlled system elements 4 H_ : Feedback elements mt w am b Fig. 1.9 Basic structure of a feedbaclactual physical quantity. For example in the heating system of Fig. 1.8, y, is srature, but this is not an actual measurable temperature; it exists only * printed on a dial. In fact, the actual physical input to the control rence input r for the heating system. The resents conversion oft command signal othe ference ipa ; can, when necessary, perform a more sophisticated function | contaminated by high-frequency noise). ymibo| for the error detector which compares the ignal €=(r—b). The feedback elements b o tion elements block, D, produces the control signal u ‘ontroller is generally thought of ignal and is also responsible fora The co about the d (manipulated variable m) to infl the ‘spuscle” of a conta “under contro! function ele Disturbance in information variabl n is mainly with the i itr (derived from the co For example, in the proportional control la ¢ function elements block, the control function elem n in Chapter 7). However, the’ fe will, therefore, not roduced in Section 1.3.| THE GENESIS AND ESSENCE OF THE 1.2 | FEEDBACK CONTROL THEORY ——_—_____—_—_—_ From the simple and familiar examples of automatic control systems presented in Section 1-1, conclusions can be derived regarding the characteristics common to these systems, notwiths their manifestable heterogeneity. Automatic control systems share the following general fea 1. The principle of operation of the system uses the feedback idea, i.e., there exists a elo error —> manipulated variable —> controlled variable > error. 7 2. The primary objective of the system is error self-nulling, Inall practical control systems, the objective of error self-nulling is required to be prescribed accuracy (steady-state errors, speed of response, etc.). A demand on shortly se accuracy and adequate stability margin is the principal accomplishment of the, theory. Detailed quantitative study of the effects of feedback on control system perfor taken up in Chapter 4. The objective of the brief qualitative study in this section is to i principles of the feedback control theory. i A brief consideration of the origins of the feedback control theory will be in order feedback control systems have been identified dating back to the third century BC. Kt in Alexandria, is credited with building a self-regulating flow device for use in water Watt of Glasgow developed his now famous flyball governor for this fiyball governor is still used for speed control applications. Analytical tools for control problems were first developed by J CN lized a linear differential equation approach and derived stability In 1876, Edward John Routh was able to determine the stability con solution to the stability problem was also obtained by Adolf Hurwit Hurwitz were unaware of each other’s work. In 1911, equivalen was established From 1900 to 1940, significant developments oceurred in lary tics, chemical industries, and electronics. The development of the resulted in many analytical techniques for the design of feedl One of the key problems in electronics, especially since its was to de Scattering of characteristics in the mass-produced electronic teristics taking place in time (¢.g., due to emission loss), negative feedback amplifier as a solution to this problem. Fi ability is a notion that describes whether the system will be al 'n anonrigorous sense, a system is said to be unstable if its output When the input is within bounds. éAmplifier > with gain G 9 (a) Fig. 1.10 Amplifier with feedback A potential-dividing resistor delivers a part, Hy, of the amplifier’s outpy input, where it is subtracted from input voltage r. Figure 1.10b gives the system. It follows from this diagram that y= G(r-Hy) ‘yea The closed-loop gain: Ms2-— St a. G This equation, specifying the relationship between the closed-loop. and the feedback gain H, is crucial for the feedback theory which studies of Harry Nyquist and Hendrik W Bode in the 1930s) and wi widely fostered frequency methods of analysis. The most important which determines feedback performance is that Vie ee 1 i fore < 3 computer circuit Pr trigger control circuit ig. 1.14 Variable speed de drive Firing angle of the SCR controls the average armature current which: de motor. The average armature current (speed) increases as the trigger of firing of the SCR, and the average armature current (speed) the SCR is increased In the scheme of Fig, 1.14, the reference voltage which cor moto of the motor. The occurrence of the error in speed causes an error circuit which controls the firing angle of the SCR in a direction to processing of the error signal (calculation of the control signal) is b a steady-state error exists between the actual speed and the desired) of the steady-state error can be eliminated by generating the controll component proportional to the error signal, and the other pi This proportional-integral control law will be examined in detail ini A point which needs a special mention here is that the presence for using feedback control. Without disturbances, there is probal Commonly occurring disturbances in the control system examples 1. Environment temperature in heating system (Example 1.2); 2. Wind power in servomechanism for steering of antennal(E3 3. Variations in load on de motor in variable speed de drive ler disturbances which affect the performance of control syster Oth 1. Uncertainty in our estimate of the hardware parameters of the the control system (the estimates of the parameters are used fOr Changes in the parameters taking place in time (due to wear, ageing, é1 In systems controlling mechanical motions, mechanical vibrations i components (noise) in the output signals of position and speed sensors: Co fluid in tank reactors, gives rise to noise in output signals of temperature 5[As we shall see later in this book, feedback control systems (i) greatly reduce the effect on the controlled variable of all external disturbances except those associated with the sensor (measurement system), and (ii) are tolerant of variations in hardware parameters other than those of the sensors. The feedback signal coming from the sensor of controlled variable contains useful information related to disturbances (external disturbances and variations in hardware parameters), which is usually ofa relatively low frequency. It may often include high frequency ‘noise’ introduced by the measure= ment sensors. Such noise signals are too fast for the control system to correct; low-pass filtering is needed to allow good control performance. fs FEEDFORWARD-FEEDBACK CONTROL 2 STRUCTURE This section presents the principle of one of the most profitable control schemes: feedfo earlier sections, we studied the principle of feedback control; a very simple technique that ¢ any disturbance affecting the controlled variable. When a disturbance enters the proces variable deviates from its desired value (setpoint) and, on sensing the error, the manipulates the process input in a way favoring the dissolution of error. The main back control system is that in order for it to compensate for disturbances, the conte first deviate from its desired value. Feedback control acts upon an error between controlled variable, This means that once a disturbance enters a process, it must propa cess and force the controlled variable to deviate from the setpoint before corrective & manipulated variable to maintain the controlled variable at its desired tation of such a control results in undisturbed controlled variable. The d does not generally require any specialized control theory; basic cases. Success of disturbance feedforward control schemes measure the disturbance, and (ii) estimate the effect of the disturbance on the controlled variable, ‘onsider an example of disturbance feedforward. Example 1.5 A tubular heat exchanger hematic diagram of a tubular heat exchanger is shown in) inside the tubes of the heat exchanger and is heated by the steam’ (Fig. 1.15a), The objective is to control the outlet temperature @ ariations’ in the process fluid flow Q and its inlet temperature cinperature in the presence of disturbances can be accompli as shown in Fig, 1.15b. The flow controller consists of an electrom {In a typical application, the heat exchanger is a part of the oy caused by other processing operations taking place in the system ( ‘ation system wherein the heat exchanger is a subsystem), |Process fluid Flow controller Temperature Steam ane Process fluid 0, —— 1 Condensate (b) ©, @) and the setpoint voltage corresponding to the desired 4 pneumatic valve®, whose essential features are shown in the valve comprises of a diaphragm whose position is set by the balane a spring and the controller output pressure on the diaphragm. actuate: “ By far, the most common form of valve used in process control is the pReumat can provide large power output. Pneumatic systems use a readily available working me air supplies, , and are commonly used in process industry because of their explosion=p simplicity and ease of maintenance:In the control scheme of Fig. 1.15b, the feedback controller, which manipulates the heat input to peat exchanger, will not act until an error has developed. If the system involves large time lags, i take some time before any corrective action takes place. Sait If feedforward control is provided in such a system, then as soon as a change in Q a acorrective measure will be taken simultaneously by manipulating the heat input to. Figure 1.16 shows the disturbance-feedforward control scheme. Feedforward control calculations Process fluid Q, Fig. 1.16 Disturbance compensated open-loop system Feedforward control can minimize the: ‘cgusenteroe However, since itis an are limitations to its functional accuracy. Feedforward control will not cancel able disturbances. It is, therefore, necessary that a feedforward control systen as shown in Fig. 1.17. Essentially, the feedforward control minimizes the measurable disturbances, while the feedback loop compensates for any i Flow sensor 4 |—| valve ControllerControl Systems: Principles and Design Command. compensator Command) variable Fig. 1.18 A control scheme employing command feed} is augmented by command-compensator to produce improved performance. been used in several ways for servo applications. In the next section, we will Several manipulated inputs available to provide this control. Sometimes, the one input affects primarily one output and has only weak effect on the other. itis possible to ignore weak interactions (coupling) and design controllers one input affects only one output. Input-output pairing to minimize the eff cation of SISO control schemes to obtain separate controllers for each acceptable performance. This, in fact, amounts to considering the multi of an appropriate number of separate SISO systems. Coupling effects are to the separate control systems and may not cause significant degradation coupling is weak. We have already discussed examples of the multiva automobile driving system (Example 1.1), and the antenna stabilization. resulting multivariable system as consisting of an appropriate nun design a controller for each system. 2. Design a single controller for the multivariable system, taking intera In this section, we give examples of multivariable systems with strong other important applications are difficult to present convineingly in an dynamic modeling is too complex to be accommodated. We will ‘therefore’ features of these control systems. Selected titles for detailed study are [119 Example 1.6 Distillation system The purification of materials is a prominent operation of the chemical industry. Th chemical process which does not require preliminary purification of raw materials or fin products from by-products. The towers of a modern petroleum refinery are the evidence 0 tant role played by these separations in a processing plant.Coolant Condenser} | Accumulator Vapor +] Falling ( liquid Feed Rising vapor Condensate Bottom produet is partially vaporized by heating, it is found that the newly e1 contain ammonia and water, but in proportions which at e and different from those in the original solution. The vapor nent (ammonia) and the residual liquid is richer in the less v original solution. If the vapor and the liquid are separated fi ization and condensation is continued over a number of $I very high degree of purity in the more yolatile componThe thermal efficiency of this staged operation can be greatly improved if the energy released the vapor is condensed in one stage is used to vaporize the liquid in an adjacent stage. If the required to vaporize one mole (a mole of a substance is defined as the amount of substance whose m numerically equals its molecular weight) of liquid is equal to the heat that is released when one of vapor conde ind is independent of composition, then it needed only at the ends of the cascade, which is rich in the more volatile component. Let us now consider a simple control scheme for the distill In the distillation column, the necessary heat transfer is accomplished by the vapor and liquid by trays (perforated plates) which are designed so that vapor: through the liquid as illustrated in Fig, 1.19. The feed (multicomponent, in ge vertical cascade of stages close to its center, The feed mixes with the * at every stage is stripped of the more volatile component by the rising bottom by partial vaporization of the bottom liquid in the reboiler. The rich in the less volatile component and is removed. The rising vapor gets: more volatile component. It is condensed and stored in the reflux the top of the tower is the reflux and the material removed from the reflux it is evident that a heat source an lation system, the control are to maintain two product compositions, xp (overhead (di and_x, (bottom product composition), at their setpoints. The three the vapor balance by controlling the column pressure and the in the reftux accumulator and the column bottom. ables do not affect the operation of the column directly; the treated as independent SISO systems. Condenser Composition controller Fig. 1.20 Distillation-column reflux In the control scheme shown in Figs 1.20-1.21, the distillate rate is m of the reflux accumulator, and the bottom rate is manipulated to control the liquid b level coComposition setpoint Xp5 Level sensor Compo: c e ql controller ontroller Bottom product Fig. 1.21 Distillation-column reboiler Ina column, the pressure is determined solely by the heat balance. If heat is at a higher rate than it is removed from the condenser, then the pressure in the Column pressure can be controlled by manipulating the reboiler heating rate or rate. We will me that the pressure controller manipulates the condenser ci a non-interacting control loop (not shown in the figures). This reduces the problem to a two-by-two multivariable control problem: th are steam rate Qy and reflux rate Q,, and the controlled variables are the rnd the distillate composition x, Figure 1.22 shows a block diagram two-input, two-output system. Changes in the input variable O, cause re and xp. Similarly, input variable Q, affects x, as well as xp, xp!Qr relation relation, Fig. 1.22 Block diagram representation of the multivariablExample 1.7 Aircraft stability and control Aircraft control is one of the most important applications through which have developed. In aireraft contol systems, the ‘seam! aels as. ‘plan ie be controlled. Main parts of an airframe include (Fig, 1,23): ce 1. Fuselage or mainbody, Wings with ailerons, 3. Horizontal tail plane with elevator, and 4. Fin or vertical tail plane with rudder. Fuselage Fig. 1.23 Control surfaces of an ai ‘or and rudder are the control ‘ol surface is a movable (hinged) a portion of the fixed aerody- ace (wing, horizontal tail, vertical tail) to vary the lift coefficient 24 shows aileron deflec- n angle 6, is defined as the mean ar displacements of the two aile- 1 two controls which can be manip- i the elevator deflection 6, 4G y lity and control of a rigid ary to separate the six degrees f freedom into two groups of three each, One: group d t tudinal dynamics of the raft and comprises the tran along the OX- and OZ-axe ional motions 4 nd the pitching motion, The other group dese remaining three degrees of freedom, namely, the axis, and the ng and rolling motions. The elevator control forces and the thi motion, whereas the aileron and rudder primarily affect the lateral motions 18 the preliminary design work, the coupling of the lateral motion into longitudinal raft and comprises th dindistribution on a surface, which in the lift coefficient causes & re of In general, a control functions by causing @ change in the pressure results in a change in the lift coefficient of the surface, This change i ‘hange in the moment balance of the aeroplane and results in the angular moment about one or mo axes. The principal control actions of the three control surfaces are: th 1. The rudder controls yawing, 2. The ailerons control rollin, 3, The elevator controls pitehin; e,, motion about the OZ e., motion about the OY-axis, and i.e, motion about the OY-axis. the actual positioning of the control surfaces is obtained by electric or hydraulic motors which controlled by the pilot, Later in this book, we shall analyze aircraft motions (SISO systems) under assumptions: neglect many details but preserve the essential nature of the problem. Mathematical mod aircraft dynamics is beyond the scope of this book, References [14~16] give a good treati subject. Example 1.8 — Robotic control system Machines that automatically load and unload, cut, weld or cast, are used by the indt obtain accuracy, safety, economy, and productivity, Programable computers integrated that often substitute for human labor in specific repeated tasks, are the modern robots.’ $ iss, and ban A robot manipulator is made of several links connected usually in series by the, arm. A link its joint. When the actuator of a joint causes rotational motion, the link is ea actuator produces translational motion, the link is called prismatic, A grippe 2 hand or an end effector, is attached to the arm by means of wrist joints is to orient the end effector properly. The motion of a manipulator end effector is caused by the movements | ions of the joints determine the configuration of the arm, which place: cation in the environment. The motion of the joints produced by the and orientation of the end effector at any time. Transducers such as eneo used to provide information for determining the position and orientati control the manipulator motion. “ The set of all points that can be reached by the end effector of a coordinates, which represent three degrees-of-freedom, Similarly, a effector is determined by three independent variables. Thus, six i describe the position and orientation of the end effector, Commer six joints, giving six degrees-of-freedom. However, robots with I cific tasks, are also available. Figure 1.25 shows a picture of th Manipulator for Assembly), a revolute manipulator, "The word ‘robot’ seems to originate from the Czech word robot1g Dosign Control Systems: Principles 27% Waist rotation Shoulder rotation The objective in providing the position control for a robot m ate controller for the robot motors, so that the position and the the desired trajectory with no errors even in the presence of d a highly complex system—each link of the robot arm has vg on all the other links as the configuration of the robot multilink manipulator describe a MIMO system; the e interactions of the joints (This MIMO system cannot be number of separate SISO systems), For a six-joint robot, the position vector @ in Fig. 1.26) are the controlled outputs, and actuator torques 1; i=, ..., 6 (represented by the input vector u in Fig. 1.26) are the manipulated vari- ables. 8, is the position vector correspond- ing to the desired trajectory. Perfect tracking can be achieved by a scheme employing command feedforward, Let @,(1) be the input to a feedforward con. troller with dynamics which are inverse of the dynamics of the manipulator, When the manipulator is driven by the output of such a controller, we get @(0) = @,(i), 7MCC —=$S rhe command-tracking based on this imerse dynamics concept can be perfect only if we can exactly getermine the dynamic model of the manipulator. The robotic system is too complex to allow exact identifica ystem is subject to unmeasurable disturbances. Undesirable d jranipulator from the desired trajectory can be corrected by mea gram showing the total control system (primary controller (fee {feedback)) is displayed in Fig, 1.26, The problem of contro! of the multilink manipulator is too compl tory text. To bring robot application into our range, we will consider th jink manipulator (SISO system), The problem of control of a two-link m in the companion book [155]. Selected titles for further reading are [17-19] tion of the dynamic model. Also, the leviations of the actual motion of the ng of feedback control. A block dia- forward) and secondary controller lex to be included in this introduc~ .e problem of control of a single~ anipulator has been examined CONCLUDING REMARKS Basic understanding in this chapter, we discussed open- and closed-loop control systems. castem engineering was given through some illustrative examples. Our description has tive in nature. In Chapters 2 and 3, we will build the platform for quantitative description. platform to deliberate on issues related to feedback control will follow in later chapters. What we have learnt? % + In industrial practice, a system is subjected to different commands (input) at system response (output) is required to follow these commands. We require a Co’ force the system to operate in command-following mode. A good control system that reacts to the: uncertainties (disturbances generated ff and from outside) is necessary. One way to react is to measure ‘the disturbances system, and to manipulate the system so as to cancel their effects. Such a cane possible because disturbances are usually unmeasurable. It is, therefore, n ances affect the system, measure the resulting error in system output, if nat is going wrong, and let the controller react to it. By doing this, we about the current situation, to the system, i.e., we are creating a were no uncertainties in the system and its environment, there would be n: open-loop control would be feasible—even better. edback control system consists of interconnected subsystems to ave learnt these issues through functional block diagrams of two i rol systems: (i) motion control systems with speed/position as resp ontvol systems with temperature/pressure (liquid level in a tanks) as te 1.1. Define the following terms commonly used in control systems: xt, open-loop control system, closed-loop control system ( control system, regulator system, setpoint control, set system, SISO system, MIMO system, feedforward-feedback 1.2 The introduction of feedback is mainly motivated by the ineo system to be controlled and the effects of external disturbs