BROWSE
Detail
Conference Paper
Integrity Heuristic as a Tie-Breaking for Classical Task Planning
Cited 1 time in 
Share 
Share
- Authors
- Joonmyun Cho, Hyonyoung Han
- Issue Date
- 2023-10
- Citation
- International Conference on Control, Automation and Systems (ICCAS) 2023, pp.1850-1853
- Language
- English
- Type
- Conference Paper
- Abstract
- There is a growing interest in automated planning (also known as AI planning) as a promising technique for autonomous agents, such as robots, to achieve a given goal (mission) by devising their own action plans. In AI planning as heuristic search, informative and fast heuristics are crucial for performance. Recently, AI planning systems using state-of-the-art heuristics have made remarkable progress compared to the past, mostly achieved by using a combination of multiple heuristics. In this paper, we propose a novel search heuristic called the integrity heuristic, which is based on the popular FF heuristic. The integrity heuristic is used in conjunction with the FF heuristic through a tie-breaking mechanism to improve the performance of planners. In this paper, we present the theoretical grounds for defining the integrity of states and empirically verify our integrity heuristic by implementing a greedy best-first search algorithm using the FF heuristic and our integrity heuristic, and show that the proposed integrity heuristic leads to performance improvement.
- KSP Keywords
- AI planning, Action Plans, Automated Planning, Best-first search, Breaking mechanism, Heuristic search, Planning system(CPPS), Search Algorithm(GSA), Search heuristic, Task planning, Tie-breaking