This benchmark solves planning problems incrementally. Each instance of a
planning problem can be described as:

- initial clauses I: satisfied in the initial statet t[0]
- goal clauses G: satisfied in the goal state t[i]
- universal clauses U: satisfied at every time point t[k]
- transition clauses T : satisfied at each pair of consecutive time point t[k],
t[k+1]

The instances have been generated from a subset of the benchmarks of the
Planning Competition (IPC) 2014 with help of a modified version of the planner
madagascar [1,3].

The problems are encoded using the double ended incremental encoding. Details
can be found in [2].



The software is available via GitHub [4].

Stephan Gocht
stephan.gocht@student.kit.edu
09.05.2017

[1] Planning as satisfiability: parallel plans and algorithms for plan search,
Rintanen, J.; Heljanko, K.; and Niemelä, I., 2006 Artificial Intelligence

[2] Accelerating SAT Based Planning with Incremental SAT Solving, Stephan Gocht
and Tomas Balyo, 2017 The 27th International Conference on Automated Planning
and Scheduling

[3] https://users.ics.aalto.fi/rintanen/satplan.html

[4] Latest:
https://github.com/StephanGocht/incplan

Based on Commit:
9413714018436313b606b986f373a15ba612e8f2
https://github.com/StephanGocht/incplan/tree/9413714018436313b606b986f373a15ba612e8f2

