Git Product home page Git Product logo

tsltools's Introduction

build

TSL Tools

A tool set and library for processing and synthesizing Temporal Stream Logic (TSL) specifications.

  1. Tool Overview

    1. Analyzing TSL Specifications
    2. Processing TSL Specifications
    3. Generating Control Flow Models
    4. Debugging TSL Specifications
  2. Installation

  3. Research and Documentation

  4. Contributing

Tool Overview

To run full pipeline synthesis, use the tsl synthesize command. This takes a TSL spec and outputs a controller programmed in the specified target language. For a quick test, you can run:

tsl synthesize -i test/res/specs/Heating.tsl --python

The precise usage and arguments for each tool are describe by tsl --help and tsl <subcommand> --help. Note that most subcommands will try to read some file from STDIN when they get no specific input.

Note that you will still need an LTL synthesis engine. See Installation for further instructions.

Processing TSL Specifications

The synthesis pipeline has the following passes:

  1. 'preprocess': convert a TSL specification with full syntax to a TSL specification with base syntax.
  2. 'theorize': generate assumptions for the theory that the TSL specificaiton is using.
  3. 'tlsf': under-approximate the TSL specification in TLSF.
  4. 'hoa': synthesize the TLSF specification with user-provided ltlsynt into a controller in HOA format.
  5. 'synthesize': generate controller program in target programming language.

The names of each pass also corresponds to a subcommand of the tsl executable, which outputs the artifact processed at that pass. For example, tsl preprocess takes a TSL specification with full syntax and outputs the specification in base syntax; on the other hand, tsl hoa takes the TSL specification with full syntax and outputs the synthesized controller in HOA format.

Therefore, to run the full pipeline, use the tsl synthesize command.

Debugging TSL Specifications

  • UPCOMING: tslplay allows to play against a environment strategy (system strategy) as the system (environment) interactively. tslplay shows why some options are not available to the user according to the respective specification helping to understand why some specification are unrealizable.
  • tsl coregen generate so called TSL unrealizability cores, i.e. the minimal amount of guarantees of some specification that render it unrealizable. Can be run with e.g. tsl coregen -i spec.tsl.
  • tsl minrealizable generate so called minimal assumption cores, i.e. the minimal amount of assumptions of some specification that render it realizable.

Installation

We use the Haskell Tool Stack for building. The tool automatically pulls the required version of the Glasgow Haskell Compiler (GHC) and all required dependencies. Note that by using stack, the installation does not interfere with any system installation. After stack is installed, when in the main project directory, you can try the tsl command with:

stack run -- tsl SUBCOMMAND

If you want to install the tsl command to you system, you can use:

stack install
# OR, if you want to specify a custom installation destination:
stack install --local-bin-path SOMEPATH

You will also need to install an LTL synthesis engine. Recommended options are ltlsynt or Strix. ltlsynt is packaged with spot.

Research and Documentation

If you want to contribute please refer to CONTRIBUTING.

tsltools's People

Contributors

wonhyukchoi avatar stemar94 avatar santolucito avatar dc3347 avatar kleinreact avatar phheim avatar leoqiao18 avatar rheakk avatar shmublu avatar lucasoethe avatar ravenrothkopf avatar nupurd89 avatar mayuridev avatar leyicui-angel avatar ggeier avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.