Component-Based Design of Embedded Control Systems

Transcription

1 Component-Based Design of Embedded Control Systems Edward A. Lee & Jie Liu UC Berkeley with thanks to the entire Berkeley and Boeing SEC teams SEC PI Meeting Annapolis, May 8-9, 2001

2 Precise Mode Change Problem thread or process thread or process thread or process How do you get the processes to a quiescent state to take a mode change? SEC, Annapolis, 2

3 Components and their Relationships Entity Port connection Relation Link Link connection Link Port Entity connection Port Entity An abstract syntax - clustered graphs - is well suited to a wide variety of component-based modeling strategies, ranging from state machines to process networks. SEC, Annapolis, 3

4 Actor View of Producer/Consumer Components Basic Transport: send(0,t) receiver.put(t) get(0) Producer/consumer styles: E1 P1 R1 IOPort Actor P2 IORelation E2 token t Receiver (inside port) continuous-time dataflow discrete events synchronous time-driven publish/subscribe SEC, Annapolis, 4

5 A Laboratory for Exploring Models of Computation Ptolemy II Java based, network integrated A realization of a model of computation is called a domain. Multiple domains can be mixed hierarchically in the same model. SEC, Annapolis, 5

6 Basic Object Model for Executable Components «Interface» Executable «Interface» Actor +fire() +initialize() +postfire() : boolean +prefire() : boolean +preinitialize() +stopfire() +terminate() +wrapup() +getdirector() : Director +getexecutivedirector() : Director +getmanager() : Manager +inputportlist() : List +newreceiver() : Receiver +outputportlist() : List ComponentEntity 0..n 0..1 CompositeEntity CompositeActor Director AtomicActor SEC, Annapolis, 6

7 Abstract Semantics How Components Interact flow of control Initialization Execution Finalization communication Structure of signals Send/receive protocols SEC, Annapolis, 7

8 Abstract Semantics How Components Interact flow of control Initialization Execution Finalization communication initialize() Structure of signals Send/receive protocols preinitialize() declare static information, like type constraints, scheduling properties, temporal properties, structural elaboration initialize variables SEC, Annapolis, 8

9 Abstract Semantics How Components Interact flow of control Initialization Execution Finalization iterate() communication Structure of signals Send/receive protocols SEC, Annapolis, 9

10 Abstract Semantics How Components Interact flow of control Initialization Execution Finalization iterate() prefire() fire() postfire() communication Structure of signals Send/receive protocols stopfire() SEC, Annapolis, 10

11 The Key Action Methods Prefire() obtain required resources may read inputs may start computations returns a boolean indicating readiness Fire() produces results Postfire() commits state updates (transactional) StopFire() request premature termination All of these are atomic (non-preemptible) SEC, Annapolis, 11

12 This Abstract Semantics has Worked For Continuous-time models Finite state machines Dataflow Discrete-event systems Synchronous/reactive systems Time-driven models (Giotto) Hybrid systems Can we make it work for priority-driven multitasking (RTOS style)? SEC, Annapolis, 12

13 Benefits Composable semantics arbitrarily deep hierarchies heterogeneous hierarchies controller plant Precise mode switching task1 task2 actuator dynamics sensor nest FSMs with anything else TTA TTA Hierarchical, heterogeneous, system-level model SEC, Annapolis, 13

14 RTOS Domain Objective: understand and improve OCP semantics support priority-driven preemptive scheduling use atomic execution, to get composability solve the precise mode change problem Solution: Atomic execution when possible Façade to long-running processes when not SEC, Annapolis, 14

15 Atomic Façade to Long-Running Dispatcher Computations Each component defines the interaction between the atomic façade and the longrunning process. There are several useful patterns: allow task to complete enforce declared timing anytime computation transactional Declared time run other atomic and non-atomic operations prefire() ready priority-driven multitasking fire() postfire() continue Atomic Facade start join produce outputs commit state Operation Task SEC, Annapolis, 15

16 RTOS Domain Implementation priority executiontime RT-Q (clock, 2.0) (clock, 1.0) (actor, output time) OS-Q (T3, p3, t3) (T1, p2, t2) (T1, p1, t1) (task, priority, remaining processing time) SEC, Annapolis, 16

17 Example: two simple tasks nonpreemptive preemptive SEC, Annapolis, 17

18 Inter-domain example: shared-resource controllers plant1 plant2 computer controller1 controller2 SEC, Annapolis, 18

19 Background process example: Data acquisition and processing atomic background processes SEC, Annapolis, 19

20 What a Modal Control System Might Look Like RTOS model RTOS model RTOS model SEC, Annapolis, 20

21 Conclusion Systematic, principled, real-time, heterogeneous, hierarchical composition of: Processes and/or threads Finite automata (mode controllers) Other models of computation Continuous-time models Dataflow models The key is the abstract semantics of Ptolemy II, which defines hierarchical heterogeneous composition of models of computation. SEC, Annapolis, 21