I am interested in many practical and theoretical aspects of AI planning and related tasks like plan explanation, plan recognition, plan repair and plan verification.
These are my main lines of research:
Expressivity of Planning Formalisms – I have studied how complex structures in solutions can become for common planning formalisms
(see ECAI'14 and
ICAPS'16a). A comprehensive discussion can be found in my dissertation thesis.
Hierarchical Planning – I have been working on solvers for HTN planning based on heuristic search
(see e.g. ICAPS'18 or
JAIR'20),
on translations to propositional logic
(see e.g. AAAI'19)
and classical planning
(see e.g. ICAPS'16b or
ICAPS'21).
If you are interested in hierarchical planning, check out our survey on the topic (IJCAI'19).
Lifted Planning – I have been working on lifted heuristics
(see IJCAI'21 or ECAI'23),
lifted landmark generation
(see IJCAI'22),
and solvers based on a translation from lifted planning to propositional logic
(see ICAPS'22a).
Planning and Learning – I have been working on Deep Reinforcement Learning
(see e.g. ACM TOMACS'23),
and on using techniques from planning to gain trust in learned action policies
(see ICAPS'22b).
I am one of the main developers of several state-of-the-art planning systems:
Our PANDA system is a framework with various functionalities in the context of hierarchical planning: grounding and preprocessing, various solvers, and problem compilations e.g. of plan and goal recognition problems to planning problems or of plan verification problems to planning problems. You can find the software here, and an article summarizing the main functionality here.
My TOAD system is a translation-based HTN solver that translates HTN planning problems to classical planning problems and uses approximation instead of bounding to overcome the differences in expressivity. The approach is inspired my work on expressivity, it is described in this paper. You can find the software here.
Our LiSAT system is a translation-based solver for lifted classical planning that translates lifted classical planning problems to SAT problems in propositional logic. The approach is described here, you can find the software here.
