Differences

This shows you the differences between two versions of the page.

--- software [2019/04/04 15:18] – admin
+++ software [2025/01/13 11:08] (current) – [SibylSAT Planner] admin
@@ Line 1: / Line 1: @@
 ====== Software ======
-  * [[http://pddl4j.com|PDDL4J v3.8]] (Planning Domain Description Language for Java)} is an open source library under LGPL license whose purpose of PDDL4J is to facilitate the development of JAVA tools for Automated Planning based on PDDL language (Planning Domain Description Language).
-  * [[https://github.com/pellierd/pddl4j_rospy|PDDL4J-ROSPY v1.0.0]] is a ROS python module to interface PDDL4J with ROS python environment and use in PDDL4J on robotics plateforms.
+~~NOTOC~~
-  * Frontal Object Perception is a prototype designed to perceive mobile objects in front a car using several sensors embedded on this car.
-  * Moving Objects Classification is a prototype designed to classify mobile objects in front a car using several sensors embedded on this car.
+===== PDDL4J v4.0 =====
+###
+**[[http://pddl4j.imag.fr|PDDL4J v4.0]]** (Planning Domain Description Language for Java is a software library under LGPL license embedding Artificial Intelligence algorithms to find solutions for planning problems, that is to say time organized actions to achieve a goal. Solutions to planning problems are “todo lists” named plan representing operational features of actions like who, how, where, when and what to do. The code of the library is available on [[https://github.com/pellierd/pddl4j|GitHub]]. More details are available in :
+   * D. Pellier & H. Fiorino (2017) [[https://hal.science/hal-04728835v1|PDDL4J: a planning domain description library for Java]], Journal of Experimental & Theoretical Artificial Intelligence, 30:1, 143-176
+###
+===== SibylSAT Planner =====
+###
+**[[https://github.com/gaspard-quenard/sibylsat|SibylSat]]** is an incremental SAT-based planner for totally-ordered HTN planning problems based on a fork of the Lilotane planner. Like lilotane and other current SAT-based solver, SibylSat adheres to a standard procedure of alternating between expanding the search space, encoding it into a SAT formula, and invoking a SAT solver to find a solution plan. However, it differ from the other SAT-based planner in that it uses a greedy best first search to explore the search space instead of the usual breadth-first search.
+  * Gaspard Quenard, Damien Pellier, Humbert Fiorino: [[https://cnrs.hal.science/LIG_SIC_MARVIN/hal-04764696v1|SibylSat: Using SAT as an Oracle to Perform a Greedy Search on TOHTN Planning]]. ECAI 2024: 4157-4164
+###