Pavel Surynek's Academic Page | Software

Software


This page contains several software projects which I developed myself or in cooperation with my colleagues.

Currently, I am developing a new full-text search engine called yeRCH. It is intended to serve as a research and development platform for web related artificial intelligence projects. An on-line demo is now powering site search on this academic web. The engine consists of the standard components used by major web search engines - web crawling, indexing, and searching. New AI techniques are planned to improve user's experience. The engine is written in C++ and all the main components are developed by us.

Additionally, some notes on solving instances of the SAT problem and several AI planning benchmarks are included in this page.



Up to the present time I have placed on this page the following projects:



Several notes about solving SAT problems

I wrote a simple tool for preprocessing Boolean formula satisfaction problems. The page about the tool is accessible through the following link. Several notes and related links are also provided at the page.

  • SSAT - SAT Preprocessing Tool


  • Benchmark Planning Problems (for SPlan)

    Several STRIPS-style benchmark planning problems are available through the following link. Planning problems from this set are described using an ad-hoc xml format which I quicky developed for my experimental planning library SPlan (see software section of my web site). A planning domain, an initial situation and a goal are described for each planning problem. Actions which manipulates the planning world are not listed in problem description files. They are generated by the planner.

  • A Set of Benchmark Planning Problems (for SPlan 1.5.0 and higher)
  • The set of planning problems consists of problems for three planning domains - Dock Worker Robots, Towers of Hanoi and Refuelling Planes. Some of the problems are really hard to solve (for example dock worker robots problem number 6).