CDS 101, Lecture 130 Sep 02R. M. Murray, Caltech1CDS 101: Lecture 1Introduction to Feedback and ControlRichard M. Murray30 September 2002Goals:y Define what a control system is and learn how to recognize its main featuresy Describe what control systems do and the primary principles of controly Give an overview of CDS 101/110; describe course structure and administrationReading (available on course web page): y R. M. Murray (ed), Control in an Information Rich World, Chapter 2y “For the Spy in the Sky, New Eyes”, NY Times, June 2002.y Optional: K. J. Astrom, Control Systems Design, Chapter 130 Sep 02 R. M. Murray, Caltech CDS 2Control = Sensing + Computation + Actuation = FeedbackSenseVehicle SpeedComputeControl ActionActuateGas PedalGoals: Stability, Performance, RobustnessCDS 101, Lecture 130 Sep 02R. M. Murray, Caltech230 Sep 02 R. M. Murray, Caltech CDS 3Early Uses of Feedback and ControlWatt Governor (1788)• Regulate speed of steam engine • Reduce effects of variations in load (disturbance rejection)Feedback Amplifiers (1920s)• Laid mathematical foundations for classical control• Black’s use of negative feedback to reduce uncertainty (robustness)Pre-1700• Water clock (~300 BC, Alexandria), float valves• Egg incubator (Drebbel, 1620) - control temperatureBalls fly out as speedincreases, closing valve30 Sep 02 R. M. Murray, Caltech CDS 4Modern Engineering ApplicationsFlight Control Systemsy Modern commercial and military aircraft are “fly by wire”y Autoland systems, unmanned aerial vehicles (UAVs) are already in placeRoboticsy High accuracy positioning for flexible manufacturingy Remote environments: space, sea, non-invasive surgery, etc.Chemical Process Controly Regulation of flow rates, temperature, concentrations, etc.y Long time scales, but only crude models of processCommunications and Networksy Amplifiers and repeatersy Congestion control of the Internety Power management for wireless communicationsAutomotivey Engine control, transmission control, cruise control, climate control, etcy Luxury sedans: 12 control instruments in 1976, 42 in 1988, 67 in 1991AND MANY MORE...CDS 101, Lecture 130 Sep 02R. M. Murray, Caltech330 Sep 02 R. M. Murray, Caltech CDS 5Other Applications of FeedbackBiological Systemsy Physiological regulation (homeostasis)y Genetic regulatory networksEnvironmental Systemsy Microbial ecosystemsy Global carbon cycleQuantum Systemsy Quantum information processingy Quantum measurementFinancial Systemsy Markets and exchangesy Supply and service chainsESE30 Sep 02 R. M. Murray, Caltech CDS 6Two Main Principles of ControlRobustness to Uncertainty through Feedbacky Feedback allows high performance in the presence of uncertaintyy Example: repeatable performance of amplifiers with 5X component variationy Key idea: accurate sensing to compare actual to desired, correction through computation and actuationDesign of Dynamics through Feedbacky Feedback allows the dynamics of a system to be modifiedy Example: stability augmentation for highly agile, unstable aircrafty Key idea: interconnection gives closed loop that modifies natural behaviorCDS 101, Lecture 130 Sep 02R. M. Murray, Caltech430 Sep 02 R. M. Murray, Caltech CDS 7Example #1: Balancing an Inverted PendulumPassive Controly Make structural modifications to change the plant dynamicsy Use this technique whenever it is a viable option: cheap, robustOpen Loop Controly Exploit knowledge of system dynamics to compute appropriate inputsy Requires very accurate model of plant dynamics in order to work wellActive (Feedback) Controly Use sensors and actuators connected by a computer to modify dynamicsy Allows uncertainty and noise to be taken into account30 Sep 02 R. M. Murray, Caltech CDS 8Example #2: Cruise Controlengine hillengine des()mv bv u uukvv=− + +=−Control System++-disturbancereferencevkbkvbkuss des wind=+++1timevelocityvdes→ 1 ask →∞→ 0 ask →∞Stability/performancey Steady state velocity approaches desired velocity as k →∞y Smooth response; no overshoot or oscillationsDisturbance rejectiony Effect of disturbances (hills) approaches zero as k →∞Robustnessy None of these results depend on the specific values of b, m, or k for ksufficiently largeCDS 101, Lecture 130 Sep 02R. M. Murray, Caltech530 Sep 02 R. M. Murray, Caltech CDS 9Example #3: Insect FlightFlight behavior in Drosophila(a) Cartoon of the adult fruit fly showing the three major sensor strictures used in flight: eyes, antennae, and halteres(detect angular rotations)(b) Example flight trajectories over a 1 meter circular arena, with and without internal targets. (c) Schematic control model of the flightsystem[from CDS Panel Report, Sec 3.4. Figure courtesy of Michael Dickinson, Caltech.]More information:y M. D. Dickinson, Solving the mystery of insect flight, Scientific American, June 2001y CDS 101 seminar : Friday, 11 Oct 02Sensing:Actuation:Computation:Effect:antennae, eye, halterswings, legs, bodybrain, nervous systemrobust flight in a variety of environments; fault tolerance30 Sep 02 R. M. Murray, Caltech CDS 10Example #4: Congestion Control and the InternetTransmission control protocol (TCP)y Source: send packet to destinationy Source: resend packet if no ACK rec’dy Destination: ACK received packetsy Destination: reassemble packetsy Source: Adjust rate based on loss rateInternet router operation:y Receive packet from input link; place at end of queue, if not fully Transmit packet from head of queue to next router on pathy Update route table based on link statusMore information:y www.howstuffworks.comy CDS 101 seminar: Friday, 25 Oct 02Sensing:Actuation:Computation:Effect:data, ACK packetstransmit rate, router pathssrc, dst, router processorshigh speed data transmission, tolerant of link failuresCDS 101, Lecture 130 Sep 02R. M. Murray, Caltech630 Sep 02 R. M. Murray, Caltech CDS 11Example #5: RoboCupRoboCup competitiony 5 on 5 robot teamsy Completely autonomousy Overhead vision systemy Centralized computery Radio links to robotsMovie courtesy Raff D’Andrea(Cornell)More info at www.robocup.orgSensing:Actuation:Computation:Effect:overhead camera system, wheel angle encodersmotor torques, kicker mechanismcentral computer + vehicle microcomputersagile motion in dynamic environment; world championship win30 Sep 02 R. M. Murray, Caltech CDS 12Control ToolsModelingy Input/output representations for subsystems + interconnection rulesy System identification theory and algorithms y Theory and
View Full Document
Unlocking...