aig-upf / fs-private

This is the private version of the FS planner repository
GNU General Public License v3.0
5 stars 1 forks source link
fstrips pddl planning strips

The FS Functional STRIPS planner

FS is a classical planner that works with the Functional STRIPS planning language [Geffner, 2000], a modeling language based on the quantifier-free fragment of first-order logic that includes constant, function and predicate symbols, but no variable symbols. The increased expressiveness of the Functional STRIPS language with respect to propositional languages such as standard STRIPS (which is indeed subsumed by Functional STRIPS) often results in problem encodings which are more compact, more readable, have fewer ground actions and preserve the structural properties of the problem in a manner which allows the derivation of more effective heuristics.

Along with the core of the Functional STRIPS language, the FS planner supports certain extensions which are useful both from the expressive and the computational point of view. These include existential quantification, state constraints, a fairly large library of global constraints, and the possibility of using externally-defined symbols and built-in arithmetic symbols.

This documentation covers a number of practical issues related to the use of the FS planner. The planner, however, has been used and described in a number of academic publications that can be found here, the most recent of which are [Francès and Geffner, 2015] and [Francès and Geffner, 2016a] and [Francès and Geffner, 2016b].

  1. Installation
  2. Usage
  3. Credits
  4. References

Installation

The easiest way to use the planner is by manually compiling the planner source code. Alternatively, you can build and/or use a ready-to-use image in some of the containerization solutions that we support.

Usage

You can find a high-level overview of the planner usage options here

Credits

The FS planner is partially built upon the Lightweight Automated Planning Toolkit and the PDDL parser from the Fast Downward distribution.

References