EECS 144/244. Fundamental Algorithms for System Modeling, Analysis, and Optimization. Lecture 1: Introduction, Systems

Transcription

1 EECS 144/244 Fundamental Algorithms for System Modeling, Analysis, and Optimization Lecture 1: Introduction, Systems Stavros Tripakis UC Berkeley Spring

2 Computers as parts of Systems ~98% of the world s processors are not in PCs but are embedded

3 Embedded, Cyber-Physical Systems Computers (HW+SW) embedded in a physical world Cyber = computers (literally to govern ) Physical = the rest Typically in a closed-loop (feedback) control configuration (often there are many distributed controllers) controller plant Amazing progress in sensors (and actuators, and networks, and computing power, and software) have given birth to the field 3

4 Computers can be dangerous This has been known for some time now... 4

5 Computers can be dangerous An important concern today:... 5

6 Computers can be extremely beneficial! Smart cars, roads, buildings, power grid, cities,

7 Smart cars on smart roads PATH project (Partners for Advanced Transportation TecHnology) of UC Berkeley: Demo'97: 7

8 Smart / green buildings Source: Steven Chu (US Secretary of Energy) Commercial: includes offices, schools, hospitals,... 8

9 Smart / green buildings Source: 1 Quad ~ 300 TeraWh Commercial: includes offices, schools, hospitals,... 9

10 Health care Pacemakers: 10

11 Health care Robotic-assisted beating heart surgery the surgeon views the surgical scene on a video display and operates on the heart as if it were stationary while the robotic system actively compensates for the motion of the heart [Bebek'08] 11

12 Computers as parts of Complex Systems a premium car today has: - ~80 computers (ECUs Electronic Control Units) - ~100 million lines of code - ~2km of wiring (CAN bus, other networks )

13 Automotive Computer-controlled automotive systems: Source: (image items are clickable) 13

14 Automotive E.g., transmission control: Source: (image items are clickable) 14

15 This course How can we design embedded, cyber-physical systems (CPS)? 3 elements: Systems Models Algorithms System-oriented view: look at the system as a whole, then focus on its parts Model-based design Computer-aided design techniques 15

16 INTRODUCTION: SYSTEMS 16

17 What is a system? EVERYTHING! A more useful definition? 17

18 System: definition Something that has: State Dynamics: rules that govern the evolution of the state in time 18

19 System: definition Something that has: State Dynamics: rules that govern the evolution of the state in time It may also have: Inputs: they influence how system evolves Outputs: this is what we observe 19

20 Example: digital circuits Digital circuit: State:??? Dynamics:??? 20

21 Example: digital circuits Digital circuit: State: value of every register, memory element Dynamics: Defined by the combinational part (logical gates) Time: discrete, or logical (ticks of the clock) 21

22 Example: digital circuits Systems vs. models clock System (the real circuit) Model (a finite-state machine) To reason about systems (analyze, make predictions, prove things,...), we need mathematical models 22

23 Example: digital circuits Systems vs. models clock System (the real circuit) node Circuit () returns (Output: bool); let Output = false -> not pre Output; tel Different models (finite-state machines) Different representations (languages, syntaxes) of the same underlying mathematical model 23

24 Example: digital circuits Digital circuit as a system: State: value of every register, memory element Dynamics: Defined by the combinational part (logical gates) Time: discrete, or logical (ticks of the clock) Or: State: all currents and voltages at all transistors at a given time t Dynamics: physics of electronic circuits (differential algebraic equations) Different levels of abstraction 24

25 25

26 Multi-paradigm modeling Different representations (languages, syntaxes) of the same underlying formalism. Different modeling formalisms often needed to describe the same system, e.g., at different levels of abstraction. Different modeling formalisms often needed to describe different parts of the system (subsystems). Multi-paradigm modeling: being able to deal with different types of models for the same system. 26

27 Example: plant + controller Discrete-time + Continuous-time models 27

28 Classes of models (and systems) surveyed in this course Discrete models: state machines, transition systems,... Breadth vs. depth: in-depth discussion in specialized courses Timed models: discrete-event systems, timed automata,... Continuous models: differential equations: ODEs, DAEs, hybrid,... Dataflow models: process networks, synchronous dataflow Stochastic/probabilistic models: Markov chains, Markov decision processes,... Advanced topics: combining models, multi-paradigm, multi-view modeling 28

29 We will also look at application domains Discrete models: state machines, transition systems,... Digital circuits Timed models: discrete-event systems, timed automata,... Control/embedded software Continuous models: differential equations: ODEs, DAEs,... Dataflow models: process networks, synchronous dataflow Stochastic/probabilistic models: Markov chains, Markov decision processes,... Advanced topics: combining models, multiparadigm, multi-view modeling... Signal-processing Bio 29

30 Logistics Instructor: Dr. Stavros Tripakis My office: 545Q Cory Hall (5th floor, DOP Center, card-key protected, phone) Office hours: 11am-noon Fridays, room: 529 Cory. Or appointment (stavros eecs) Lectures / discussions Wednesdays Fridays, noon 2pm: 3 hours lecture + 1 hour discussion / week (flexible) Room: 540A/B Cory Hall, 5th floor Except: Wed Feb 13 and Wed Apr 10 (classroom to be announced) Credits: 4 Grading: Based on: class participation, homeworks, midterms, final team project More details coming up Interaction is greatly appreciated! 30

31 Reading, web site Reading Papers to be distributed in class / web site Free online textbook: Other textbooks (list to be populated on web site): Clarke, Grumberg, Peled Model Checking Francois Cellier Continuous System Modeling and Cellier-Kofman Continuous System Simulation Web site: (currently previous version of the course, to be updated) bspace? prulu? 31

32 Round of introductions 32

33 Quiz Express the following in your favorite mathematical formalism: You can fool some people sometimes You can fool some of the people all of the time You can fool some people sometimes but you can't fool all the people all the time [Bob Marley] You can fool some of the people all of the time, and all of the people some of the time, but you can not fool all of the people all of the time [Abraham Lincoln] 33

34 Back to the definition of system Many kinds of systems: Chair Software Car = chassis + engine + computer + software +... Human body = heart + lungs + = many cells =... Weather Stock market Internet How to describe each of these systems as states + dynamics? Difficult (impossible) to describe some systems using our current definition 34

35 System: monolithic definition Something that has: State Dynamics: rules that govern the evolution of the state in time It may also have: Inputs: they influence how system evolves Outputs: this is what we observe 35

36 System: compositional definition A collection of subsystems that interact So, we must describe: Each subsystem (recursive definition) The interaction rules We will see yet other ways of looking at systems later 36

37 Example Bluetooth chip (Cambridge Silicon) Courtesy of J. Roychowdhury, UC Berkeley 37

38 Example Subsystems: ADC, DAC, microprocessor,, wires Interaction rules (partial list): Kirchoff's laws: At every node of the circuit: sum(all currents) = 0 Bluetooth chip (Cambridge Silicon) 38 Courtesy of J. Roychowdhury, UC Berkeley

39 39

40 Example: plant + controller Discrete-time: Synchronous composition Continuous-time: Numerical integration 40

41 Example: concurrent software Multi-thread Java program: Asynchronous composition 41

42 Biochemical Systems Reactions governing conversion of glucose to ATP (energy) and back What do these diagrams mean? metabolic network J. Roychowdhury, University of California at Berkeley Slide 42

43 Pathway: Chain of Reactions Enzyme (protein) glucose-6-phosphatase Compound catalyzes reaction Connections to other Pathways alpha-d-glucose Reaction alpha-d-glucose 6phosphate phosphohydrolase Differential Equations J. Roychowdhury, University of California at Berkeley Slide 43

44 Metabolic Pathway Maps credits: expasy.ch/biomap/ J. Roychowdhury, University of California at Berkeley Slide 44

45 This course How can we design embedded, cyber-physical systems (CPS)? 3 elements: Systems Models Algorithms System-oriented view: look at the system as a whole, then focus on its parts Model-based design Computer-science techniques 45

46 Questions How can we model systems? How can we analyze the models? Analyze = reason about system properties; make predictions about the system; prove things about the systems (actually only about the models) e.g.: Will my program throw an exception? Will my concurrent program deadlock? Will my controller stabilize my system? How much energy will my circuit consume? Will this video decoder play movies smoothly? Will this drug treat cancer?... Ideally we want the computer to do the analysis for us! 46

47 For each class of models, this course surveys fundamental algorithms Discrete models: state machines, transition systems, circuits,... Timed models: discrete-event systems, timed automata,... Scheduling, performance analysis (throughput, latency),... Stochastic/probabilistic models: Markov chains, Markov decision processes,... Solution of linear/nonlinear equations (Newton-Raphson), numerical simulation,... Dataflow models: process networks, synchronous dataflow Discrete-event simulation, verification, timing analysis,... Continuous models: differential equations: ODEs, DAEs,... State-space exploration, verification, logic synthesis, testing,... Steady-state analysis, Monte-Carlo simulation (static/dynamic, Gillespie),... Advanced topics: combining models, multi-paradigm, multi-view modeling 47

48 Systems: structure + behavior Structure: What the system is made of, its parts, subsystems, Some modeling languages focus on structure: e.g., UML class diagrams 48

49 Systems: structure + behavior Structure: What the system is made of, its parts, sub-systems, Some modeling languages focus on structure: e.g., UML class diagrams Behavior: What the system does The two are intertwined: cf non-monolithic definition This course focuses on behavior 49