Performance Heuristics for GR(1) Synthesis and Related Algorithms

TL;DR

This paper introduces and evaluates heuristics aimed at reducing the computational time of GR(1) synthesis algorithms, demonstrating their effectiveness on robotic and benchmark specifications.

Contribution

It presents new heuristics for GR(1) synthesis, including fixed-point detection and recycling, with comprehensive evaluation on diverse specifications.

Findings

Heuristics can significantly reduce synthesis times.

Effectiveness varies between robot specifications and benchmarks.

Some heuristics enable early detection of unrealizability.

Abstract

Reactive synthesis for the GR(1) fragment of LTL has been implemented and studied in many works. In this workshop paper we present and evaluate a list of heuristics to potentially reduce running times for GR(1) synthesis and related algorithms. The list includes early detection of fixed-points and unrealizability, fixed-point recycling, and heuristics for unrealizable core computations. We evaluate the presented heuristics on SYNTECH15, a total of 78 specifications of 6 autonomous Lego robots, written by 3rd year undergraduate computer science students in a project class we have taught, as well as on several benchmarks from the literature. The evaluation investigates not only the potential of the suggested heuristics to improve computation times, but also the difference between existing benchmarks and the robot's specifications in terms of the effectiveness of the heuristics.