Presentation is loading. Please wait.

Presentation is loading. Please wait.

PTIDES Project Overview

Similar presentations


Presentation on theme: "PTIDES Project Overview"— Presentation transcript:

1 PTIDES Project Overview
Slobodan Matic, Jia Zou Edward Lee, John Eidson ActionWebs, 09/29/10

2 Orchestrating networked computational resources with physical systems
Transportation (Air traffic control at SFO) Avionics Building Systems Telecommunications Automotive Instrumentation (Soleil Synchrotron) E-Corner, Siemens Factory automation Power generation and distribution Daimler-Chrysler Military systems: Deterministic Implementations of Event-driven Software Courtesy of General Electric Courtesy of Doug Schmidt Courtesy of Kuka Robotics Corp.

3 Current Practice: Printing Press
High-speed, high precision Speed: 1 inch/ms Precision: 0.01 inch -> Time accuracy < 10us Application aspects local (control) distributed (coordination) global (modes) Open standards (Ethernet) Synchronous, Time-Triggered IEEE time-sync protocol Bosch-Rexroth

4 For distributed cyber-physical systems,
Timing needs to be a part of the network semantics, not a side effect of the implementation. Technologies needed: Time synchronization Bounds on latency Time-aware fault isolation and recovery

5 A Programming Model for Distributed Cyber-Physical Systems
The question we address: Given a common notion of time shared to some known precision across a network, and given bounded network latencies, can we design better distributed embedded software? Our answer (PTIDES): Use discrete-event (DE) models for specification of systems, bind model time to real time only exactly where this is needed.

6 Outline Introduction PTIDES Basics and Workflow
PTIDES Runtime Framework Single CPU: Tunneling Ball Device Multiple CPU: Power Distribution Subsystem Summary

7 PTIDES Basics Programming Temporally Integrated Distributed Embedded Systems Based on Discrete-Event model of computation Event processing in time-stamp order Deterministic under simple causality conditions Fixed-point semantics Can be extended to super-dense form Timing constraints for interaction with environment

8 Model vs. Physical Time At sensors and actuators
Relate model time (τ) to physical time (t) t ≥ τ t ≤ τ τ1 do i4 model time τ1 τ4 physical time t1 t4

9 Single Processor PTIDES Example
Bounded delay between any two hardware components Bounded sensor latency (d0) t ≥ τ , t ≤ τ + do t ≤ τ τ1 do i4 τ2 model time τ2 physical time t2 e2 at i2

10 Single Processor PTIDES Example
t ≥ τ , t ≤ τ + do t ≤ τ τ1 do i4 τ2 model time τ2 physical time t2 τ2+d0 e2 safe to process if t > τ2 + do

11 Single Processor PTIDES Example
t ≥ τ , t ≤ τ + do t ≤ τ τ1 do i4 τ2 model time τ1 physical time t2 τ1+d0 e2 safe to process if t > τ2 + do

12 Distributed PTIDES Example
Local event processing decisions: Bounded communication latency (d0) Distributed platforms time-synchronized with bounded error (e) τ is safe to process if: t > τ + do2 + e - d 2 d 1 τ1 τ2 d01 Sensor Merge τ Actuator d 2 τ3 τ4 do2 Network Interface o3

13 Distributed PTIDES Example
Local event processing decisions: Bounded communication latency (d0) Distributed platforms time-synchronized with bounded error (e) τ is safe to process if: t > τ + do2 + e - d 2 d 1 d01 Sensor τ1 Merge τ Actuator d 2 τ3 τ4 do2 Network Interface o3

14 What Did We Gain? e1 = (v1, τ1) δ e1, e2, … e2 = (v2, τ2)
First Point: Ensures deterministic data outputs Merge e1 = (v1, τ1) δ e1, e2, … safe to process analysis for e safe to process analysis for e e2 = (v2, τ2) Second Point: Ensures deterministic timing delay from Sensor to Actuator t ≤ τ + do t ≤ τ i4 do τ1 τ2 Now that I defined the GES, let’s take a step back and

15 PTIDES Program Design Workflow
HW Platform PTIDES run-time

16 PTIDES Runtime Framework
Software components (actors) are “glued together” by a code generator into an executable Lightweight real-time kernel (5KB) Actors are state machines, each with an event queue and context Asynchronous delivery: producer posts to event queue Framework handles thread-safe event exchange and queuing   Run-to-completion : No blocking code No busy-waiting, semaphore, mailbox Preemptive -> Single-stack implementation Encapsulate each resource inside a dedicated active object rest of app shares the dedicated active object via events  Fast and deterministic memory allocation Event pools:  fixed-size memory blocks

17 PTIDES Scheduler Two layer execution engine
Event coordination (safe-to-process) Event scheduling (prioritize safe events) Earliest Deadline First foundation Deadline based on path from input port to actuator Schedulability analysis requires WCET of software components + event models PTIDES implementation delivers DE semantics even w/o WCET Even if deadlines missed, events are determinate PTIDES repeatable even if WCET wrong

18 PTIDES Workflow HW Platform Software Component Library Ptides Model
Code Generator PTIDES run-time Code Plant Model Network Model HW in the Loop Simulator Causality Analysis Program Analysis Schedulability Analysis Analysis Mixed Simulator

19 Schedulability Analysis
Three cases: Zero event processing time assumption (feasibility test) if P fails, P will not satisfy constraints on any hardware No resource sharing assumption (an event is processed as soon it is safe) if P fails, P may still satisfy constraints on other hardware Resource sharing (a safe event is processed according to a scheduling algorithm) if P fails, P does not satisfy this implementation (and algorithm)

20 Feasibility Test Feasibility assumptions
Zero event processing time Bounds on event communication latency (d0) If test fails, no HW satisfies constraints detect flows

21 Feasibility Test For each input port i on a sensor-actuator path

22 Single CPU - Proof of Concept
Tunneling Ball Device sense ball track disk adjust trajectory

23 Single CPU Proof of Concept
Event sequences periodic quasi-periodic sporadic

24 Tunneling Ball Device – 10 rps

25 Tunneling Ball Device – 1 rps

26 Distributed Testbed Objectives
Safe event processing Offset attribute of a port Check: real time vs model time + offset Bounded latency between hardware components Distribution: bounded-delay networking protocol Scheduling Deadlines defined by sensor – actuator model delays Single CPU: use Earliest Deadline First Distribution: local from end-to-end deadlines Expressiveness of interaction Complexity of real sensor, actuator and network interfaces Timing support External to microprocessor National Semiconductor DP83640: IEEE 1588 enabled PHY chip

27 Synchrophasor-based measurement and control
Power swing and Unstability detection

28 Experiment Diagram Grid emulator built with National Instruments PXI
‘PMU’s built with Renesas demo boards with DP83640 Ethernet bridge or 1588 boundary/transparent clock ‘SVP’ built with Renesas demo board with DP83640

29 Basic PTIDES Timing Testing
Vary phasor data independently Freq. and phases w.r.t. global time Sample voltage and current Signal processing Send phasor data Local control: on/off breaker Wireshark monitoring of network events Detect discrepancies If unstable region send on/off command

30 Single platform PTIDES approximation model

31 Multi-platform PTIDES approximation model

32 Summary The networking problem requires timing to be a correctness property rather than a quality of service PTIDES model of computation offers a possible programming model for distributed CPS Correct deterministic DE implementations Better match for reactive state-machine systems  than traditional multi-threaded     or     super loop     programming models Time stamps provides runtime information for detecting and reacting to variety of timing faults Goal: scalability, robustness, fault-tolerance


Download ppt "PTIDES Project Overview"

Similar presentations


Ads by Google