In recent years, ML-based heuristic functions for automated planning have shown increasing performance. A main challenge is the level of generalization required in planning: techniques must generalize at least across different instances of the same domain (which results in different sizes of learning input). A common approach to overcome the issue is to use graph representations as input. While GNNs are a natural choice for learning, other methods have recently been favored because they show better runtime performance and need less training data. However, existing work has so far been limited to non-hierarchical planning. We describe the first approach to learn heuristics for hierarchical planning. We extend the Instance Learning Graph – a graph structure used in non-hierarchical planning – to the new setting and show how to learn heuristic functions based on it. Since our heuristics are applicable to the lifted model, there is no need to ground it. We therefore combine it with a novel lifted HTN planning system. Like recent systems in non-hierarchical planning, it grounds the search space explored so far, but not the entire model prior to search. Our evaluation shows that our approach is competitive with the lifted systems from the literature, though the ground systems achieve higher coverage.

Recent interest in solving planning tasks, where full grounding is infeasible, has highlighted the need to compute heuristics at a lifted level. We turn our attention to the evaluation of the hᵃᵈᵈ heuristic, which is an important cornerstone in many classical planning approaches, including the best performing lifted planning approach. We show that hᵃᵈᵈ’s grounded efficiency does not extend to lifted tasks, where the computation is EXPTIME-complete. This prompts to identify tractability islands matching practical use cases. We identify two, where a lifted computation is feasible while grounding may fail: The first constraints to acyclic action schemata and bounds predicate arity. For the second case we introduce a novel computation, operating without grounding. Assuming the extraction encounters only acyclic conditions, and hᵃᵈᵈ values per subgoal are bounded, it remains tractable. (Even with unbounded predicate and action arity.) In an empirical evaluation of the new technique, we observe complementary behavior to the existing lifted forward hᵃᵈᵈ evaluation. Combining both sets a new state-of-the-art in pure-heuristic performance on the hard-to-ground benchmarks.

In many automated planning applications, action costs can be hard to specify. An example is the time needed to travel through a certain road segment, which depends on many factors, such as the current weather conditions. A natural way to address this issue is to learn to predict these parameters based on input features (e.g., weather forecasts) and use the predicted action costs in automated planning afterward. Decision-Focused Learning (DFL) has been successful in learning to predict the parameters of combinatorial optimization problems in a way that optimizes solution quality rather than prediction quality. This approach yields better results than treating prediction and optimization as separate tasks. In this paper, we investigate for the first time the challenges of implementing DFL for automated planning in order to learn to predict the action costs. There are two main challenges to overcome: (1) planning systems are called during gradient descent learning, to solve planning problems with negative action costs, which are not supported in planning. We propose novel methods for gradient computation to avoid this issue. (2) DFL requires repeated planner calls during training, which can limit the scalability of the method. We experiment with different methods approximating the optimal plan as well as an easy-to-implement caching mechanism to speed up the learning process. As the first work that addresses DFL for automated planning, we demonstrate that the proposed gradient computation consistently yields significantly better plans than predictions aimed at minimizing prediction error; and that caching can temper the computation requirements.

Testing was recently proposed as a method to gain trust in learned action policies in classical planning. Test cases in this setting are states generated by a fuzzing process that performs random walks from the initial state. A fuzzing bias attempts to bias these random walks towards policy bugs, that is, states where the policy performs sub-optimally. Prior work explored a simple fuzzing bias based on policy-trace cost. Here, we investigate this topic more deeply. We introduce three new fuzzing biases based on analyses of policy-trace shape, estimating whether a trace is close to looping back on itself, whether it contains detours, and whether its goal-distance surface does not smoothly decline. Our experiments with two kinds of neural action policies show that these new biases improve bug-finding capabilities in many cases.

Stochastic shortest path (SSP) problems are a common framework for planning under uncertainty. However, the reactive structure of their solution policies is typically not easily comprehensible by an end-user, nor do planners justify the reasons behind their choice of a particular policy over others. To strengthen confidence in the planner's decision-making, recent work in classical planning has introduced a framework for explaining to the user the possible solution space in terms of necessary trade-offs between user-provided plan properties. Here, we extend this framework to SSPs. We introduce a notion of policy properties taking into account action-outcome uncertainty. We analyze formally the computational problem of identifying the exclusion relationships between policy properties, showing that this problem is in fact harder than SSP planning in a complexity theoretical sense. We show that all the relationships can be identified through a series of heuristic searches, which, if ordered in a clever way, yields an anytime algorithm. Further, we introduce an alternative method, which leverages a connection to multi-objective probabilistic planning to move all the computational burden to a preprocessing step. Finally, we explore empirically the feasibility of the proposed explanation methodology on a range of adapted IPPC benchmarks.

The complexity of modeling planning domains is a major obstacle for making automated planning techniques more accessible, raising the demand of tools for providing modeling assistance. In particular, tools that can automatically correct errors in a planning domain are of great importance. Previous works have devoted efforts to developing such approaches for correcting classical (non-hierarchical) domains. However, no approaches exist for hierarchical planning, which is what we offer here. More specifically, our approach takes as input a flawed hierarchical domain together with a plan known to be a solution but actually contradicting the domain (due to errors in the domain) and outputs corrections to the domain that add missing actions to the domain which turn the plan into a solution. The approach achieves this by compiling the problem of finding corrections to another hierarchical planning problem.

We present an approach for translating Totally Ordered Hierarchical Task Network (HTN) planning problems to classical planning problems. While this enables the use of sophisticated classical planning systems to find solutions, we need to overcome the differences in expressiveness of these two planning formalisms. Prior work on this topic did this by translating bounded HTN problems. In contrast, we approximate them, i.e., we change the problem such that every action sequence that is a solution to the HTN problem is also a solution for the classical problem, but the latter might have more solutions. To obtain a sound overall approach, we verify solutions returned by the classical planning system to ensure that they are also solutions to the HTN problem. For translation and approximation, we use techniques introduced to approximate Context-Free Languages by using Finite Automata. We named our system Toad (Totally Ordered HTN Approximation using DFA). For a subset of HTN problems the translation is even possible without approximation. Whether or not it is necessary is decided based on the property of self-embedding, which comes also from the field of formal languages. We investigate the theoretical connection of self-embedding and tail-recursiveness, a property from the HTN literature used to identify a subclass of HTN planning problems that can be translated to classical planning, and show that it is more general. To guide the classical planner, we introduce a novel heuristic tailored towards our models. We evaluate Toad on the benchmark set of the 2020 International Planning Competition. Our evaluation shows that (1) most problems can be translated without approximation and that (2) Toad is competitive with the state of the art in HTN planning.

Lifted planning – finding plans directly on the PDDL input model – has attracted renewed attention during the last years. This avoids the process of grounding, which can become computationally prohibitive very easily. However, the main focus of recent research in this area has been on satisficing, i.e., (potentially) suboptimal planning. We present a novel heuristic for optimal lifted planning. Our basic idea is inspired by the LM-cut heuristic, which has been very successful in grounded optimal planning. Like LM-cut, we generate cut-based landmarks via back-chaining from the goal, generating cuts of partially grounded actions. However, exactly mimicking the ground formulation is not feasible, this includes computing the hmax heuristic several times for one computation of the LM-cut heuristic (which is already NP-hard to compute). We show that our heuristic is admissible and evaluate it in a cost optimal setting.

Translation-based approaches to planning allow for solving problems in complex and expressive formalisms via the means of highly efficient solvers for simpler formalisms. To be effective, these translations have to be constructed appropriately. The current existing translation of the highly expressive formalism of HTN planning into the more simple formalism of classical planning is not on par with the performance of current dedicated HTN planners. With our contributions in this paper, we close this gap: we describe new versions of the translation that reach the performance of state-of-the-art dedicated HTN planners. We present new translation techniques both for the special case of totally-ordered HTNs as well as for the general partially-ordered case. In the latter, we show that our new translation generates only linearly many actions, while the previous encoding generates and exponential number of actions.

While state-of-the-art planning systems need a grounded (propositional) task representation, the input model is provided "lifted", specifying predicates and action schemas with variables over a finite object universe. The size of the grounded model is exponential in predicate/action-schema arity, limiting applicability to cases where it is small enough. Recent work has taken up this challenge, devising an effective lifted forward search planner as basis for lifted heuristic search, as well as a variety of lifted heuristic functions based on the delete relaxation. Here we add a novel family of lifted heuristic functions, based on landmarks. We design two methods for landmark extraction in the lifted setting. The resulting heuristics exhibit performance advantages over previous heuristics in several benchmark domains. Especially the combination with lifted delete relaxation heuristics to a LAMA-style planner yields good results, beating the previous state of the art in lifted planning.

Landmarks (LMs) are state features that need to be made true or tasks that need to be contained in every solution of a planning problem. They are a valuable source of information in planning and can be exploited in various ways. LMs have been used both in classical and hierarchical planning, but while there is much work in classical planning, the techniques in hierarchical planning are less evolved. We introduce a novel LM generation method for Hierarchical Task Network (HTN) planning and show that it is sound and incomplete. We show that every complete approach is as hard as the co-class of the underlying HTN problem, i.e. coNP-hard for our setting (while our approach is in P). On a widely used benchmark set, our approach finds more than twice the number of landmarks than the approach from the literature. Though our focus is on LM generation, we show that the newly discovered landmarks bear information beneficial for solvers.

Polynomial-time heuristic functions for planning are commonplace since 20 years. But polynomial-time in which input? Almost all existing approaches are based on a grounded task representation, not on the actual PDDL input which is exponentially smaller. This limits practical applicability to cases where the grounded representation is "small enough". Previous attempts to tackle this problem for the delete relaxation leveraged symmetries to reduce the blow-up. Here we take a more radical approach, applying an additional relaxation to obtain a heuristic function that runs in time polynomial in the size of the PDDL input. Our relaxation splits the predicates into smaller predicates of fixed arity K. We show that computing a relaxed plan is still NP-hard (in PDDL input size) for K>=2, but is polynomial-time for K=1. We implement a heuristic function for K=1 and show that it can improve the state of the art on benchmarks whose grounded representation is large.

Planning problems are usually modeled using lifted representations, they specify predicates and action schemas using variables over a finite universe of objects. However, current planning systems like Fast Downward need a grounded (propositional) input model. The process of grounding might result in an exponential blowup of the model size. This limits the application of grounded planning systems in practical applications. Recent work introduced an efficient planning system for lifted heuristic search, but the work on lifted heuristics is still limited. In this extended abstract, we introduce a novel lifted heuristic based on landmarks, which we extract from the lifted problem representation. Preliminary results on a benchmark set specialized to lifted planning show that there are domains where our approach finds enough landmarks to guide the search more effective than the heuristics available.

Both search-based and translation-based planning systems usually operate on grounded representations of the problem. Planning models, however, are commonly defined using lifted description languages. Thus, planning systems usually generate a grounded representation of the lifted model as a preprocessing step. For HTN planning models, only one method to ground lifted models has been published so far. In this paper we present a new approach for grounding HTN planning problems that produces smaller groundings in a shorter timespan than the previously published method.

The research in hierarchical planning has made considerable progress in the last few years. Many recent systems do not rely on hand-tailored advice anymore to find solutions, but are supposed to be domain-independent systems that come with sophisticated solving techniques. In principle, this development would make the comparison between systems easier (because the domains are not tailored to a single system anymore) and – much more important – also the integration into other systems, because the modeling process is less tedious (due to the lack of advice) and there is no (or less) commitment to a certain planning system the model is created for. However, these advantages are destroyed by the lack of a common input language and feature set supported by the different systems. In this paper, we propose an extension to PDDL, the description language used in non-hierarchical planning, to the needs of hierarchical planning systems.

When generating technical instructions, it is often convenient to describe complex objects in the world at different levels of abstraction.A novice user might need an object explained piece by piece, while for an expert, talking about the complex object (e. g. a wall or railing) directly may be more succinct and efficient.We show how to generate building instructions at different levels of abstraction in Minecraft.We introduce the use of hierarchical planning to this end, a method from AI planning which can capture the structure of complex objects neatly.A crowdsourcing evaluation shows that the choice of abstraction level matters to users, and that an abstraction strategy which balances low-level and high-level object descriptions compares favorably to ones which don't.

In HTN planning, the hierarchy has a wide impact on solutions. First, there is (usually) no state-based goal given, the objective is given via the hierarchy. Second, it enforces actions to be in a plan. Third, planners are not allowed to add actions apart from those introduced via decomposition, i.e. via the hierarchy. However, no heuristic considers the interplay of hierarchy and actions in the plan exactly (without relaxation) because this makes heuristic calculation NP-hard even under delete relaxation. We introduce the problem class of delete- and ordering-free HTN planning as basis for novel HTN heuristics and show that its plan existence problem is still NP-complete. We then introduce heuristics based on the new class using an integer programming model to solve it.

Search methods are useful in hierarchical task network (HTN) planning to make performance less dependent on the domain knowledge provided, and to minimize plan costs. Here we investigate Monte-Carlo tree search (MCTS) as a new algorithmic alternative in HTN planning. We implement combinations of MCTS with heuristic search in PANDA. We furthermore investigate MCTS in JSHOP, to address lifted (non-grounded) planning, leveraging the fact that, in contrast to other search methods, MCTS does not require a grounded task representation. Our new methods yield coverage performance on par with the state of the art, but in addition can effectively minimize plan cost over time.

The majority of search-based HTN planning systems can be divided into those searching a space of partial plans (a plan space) and those performing progression search, i.e., that build the solution in a forward manner. So far, all HTN planners that guide the search by using heuristic functions are based on plan space search. Those systems represent the set of search nodes more effectively by maintaining a partial ordering between tasks, but they have only limited information about the current state during search. In this article, we propose the use of progression search as basis for heuristic HTN planning systems. Such systems can calculate their heuristics incorporating the current state, because it is tracked during search. Our contribution is the following: We introduce two novel progression algorithms that avoid unnecessary branching when the problem at hand is partially ordered and show that both are sound and complete. We show that defining systematicity is problematic for search in HTN planning, propose a definition, and show that it is fulfilled by one of our algorithms. Then, we introduce a method to apply arbitrary classical planning heuristics to guide the search in HTN planning. It relaxes the HTN planning model to a classical model that is only used for calculating heuristics. It is updated during search and used to create heuristic values that are used to guide the HTN search. We show that it can be used to create HTN heuristics with interesting theoretical properties like safety, goal-awareness, and admissibility. Our empirical evaluation shows that the resulting system outperforms the state of the art in search-based HTN planning.

HTN planning provides an expressive formalism to model complex application domains. It has been widely used in realworld applications. However, the development of domainindependent planning techniques for such models is still lacking behind. The need to be informed about both statetransitions and the task hierarchy makes the realisation of search-based approaches difficult, especially with unrestricted partial ordering of tasks in HTN domains. Recently, a translation of HTN planning problems into propositional logic has shown promising empirical results. Such planners benefit from a unified representation of state and hierarchy, but until now require very large formulae to represent partial order. In this paper, we introduce a novel encoding of HTN Planning as SAT. In contrast to related work, most of the reasoning on ordering relations is not left to the SAT solver, but done beforehand. This results in much smaller formulae and, as shown in our evaluation, in a planner that outperforms previous SAT-based approaches as well as the state-of-the-art in search-based HTN planning.

Over the last years, several new approaches to Hierarchical Task Network (HTN) planning have been proposed that increased the overall performance of HTN planners. However, the focus has been on agile planning - on finding a solution as quickly as possible. Little work has been done on finding optimal plans. We show how the currently best-performing approach to HTN planning - the translation into propositional logic - can be utilised to find optimal plans. Such SAT-based planners usually bound the HTN problem to a certain depth of decomposition and then translate the problem into a propositional formula. To generate optimal plans, the length of the solution has to be bounded instead of the decomposition depth. We show the relationship between these bounds and how it can be handled algorithmically. Based on this, we propose an optimal SAT-based HTN planner and show that it performs favourably on a benchmark set.

Hierarchical planning has attracted renewed interest in the last couple of years, which led to numerous novel formalisms, problem classes, and theoretical investigations. Yet it is important to differentiate between the various formalisms and problem classes, since they show -- sometimes fundamental -- differences with regard to their expressivity and computational complexity: Some of them can be regarded equivalent to non-hierarchical formalisms while others are clearly more expressive. We survey the most important hierarchical problem classes and explain their differences and similarities. We furthermore give pointers to some of the best-known planning systems capable of solving the respective problem classes.

Planning is the task of finding a sequence of actions that achieves the goal(s) of an agent. It is solved based on a model describing the environment and how to change it. There are several approaches to solve planning tasks, two of the most popular are classical planning and hierarchical planning. Solvers are often based on heuristic search, but especially regarding domain-independent heuristics, techniques in classical planning are more sophisticated. However, due to the different problem classes, it is difficult to use them in hierarchical planning. In this paper we describe how to use arbitrary classical heuristics in hierarchical planning and show that the resulting system outperforms the state of the art in hierarchical planning.

In this paper, we propose a novel SAT-based planning approach for hierarchical planning by introducing the SAT-based planner totSAT for the class of totally-ordered HTN planning problems. We use the same general approach as SAT planning for classical planning does: bound the problem, translate the problem into a formula, and if the formula is not satisfiable, increase the bound. In HTN planning, a suitable bound is the maximum depth of decomposition. We show how totally-ordered HTN planning problems can be translated into a SAT formula, given this bound. Furthermore, we have conducted an extensive empirical evaluation to compare our new planner against state-of-the-art HTN planners. It shows that our technique outperforms any of these systems.

Planning via SAT has proven to be an efficient and versatile planning technique. Its declarative nature allows for an easy integration of additional constraints and can harness the progress made in the SAT community without the need to adapt the planner. However, there has been only little attention to SAT planning for hierarchical domains. To ease encoding, existing approaches for HTN planning require additional assumptions, like non-recursiveness or totally-ordered methods. Both limit the expressiveness of HTN planning severely. We propose the first propositional encodings which are able to solve general, i.e., partially-ordered, HTN planning problems, based on a previous encoding for totally-ordered problems. The empirical evaluation of our encoding shows that it outperforms existing HTN planners significantly.

Plan-and Goal Recognition (PGR) is the task of inferring the goals and plans of an agent based on its actions. Traditional approaches in PGR are based on a plan library including pairs of plans and corresponding goals. In recent years, the field successfully exploited the performance of planning systems for PGR. The main benefits are the presence of efficient solvers and well-established, compact formalisms for behavior representation. However, the expressivity of the STRIPS planning models used so far is limited, and models in PGR are often structured in a hierarchical way. We present the approach Plan and Goal Recognition as HTN Planning that combines the expressive but still compact grammar-like HTN representation with the advantage of using unmodified, off-the-shelf planning systems for PGR. Our evaluation shows that - using our approach - current planning systems are able to handle large models with thousands of possible goals, that the approach results in high recognition rates, and that it works even when the environment is partially observable, i.e., if the observer might miss observations.

Hierarchical task network (HTN) planning is well-known for being an efficient planning approach. This is mainly due to the success of the HTN planning system SHOP2. However, its performance depends on hand-designed search control knowledge. At the time being, there are only very few domain-independent heuristics, which are designed for differing hierarchical planning formalisms. Here, we propose an admissible heuristic for standard HTN planning, which allows to find optimal solutions heuristically. It bases upon the so-called task decomposition graph (TDG), a data structure reflecting reachable parts of the task hierarchy. We show (both in theory and empirically) that rebuilding it during planning can improve heuristic accuracy thereby decreasing the explored search space. The evaluation further studies the heuristic both in terms of plan quality and coverage.

There are several formalizations for hierarchical planning. Many of them allow to specify preconditions and effects for compound tasks. They can be used, e.g., to assist during the modeling process by ensuring that the decomposition methods' plans “implement” the compound tasks' intended meaning. This is done based on so-called legality criteria that relate these preconditions and effects to the method's plans and pose further restrictions. Despite the variety of expressive hierarchical planning formalisms, most theoretical investigations are only known for standard HTN planning, where compound tasks are just names, i.e., no preconditions or effects can be specified. Thus, up to know, a direct comparison to other hierarchical planning formalisms is hardly possible and fundamental theoretical properties are yet unknown. To enable a better comparison between such formalisms (in particular with respect to their computational expressivity), we first provide a survey on the different legality criteria known from the literature. Then, we investigate the theoretical impact of these criteria for two fundamental problems to planning: plan verification and plan existence. We prove that the plan verification problem is at most NP-complete, while the plan existence problem is in the general case both semi-decidable and undecidable, independent of the demanded criteria. Finally, we discuss our theoretical findings and practical implications.

Modern technical devices are often too complex for many users to be able to use them to their full extent. Based on planning technology, we are able to provide advanced user assistance for operating technical devices. We present a system that assists a human user in setting up a complex home theater consisting of several HiFi devices. For a human user, the task is rather challenging due to a large number of different ports of the devices and the variety of available cables. The system supports the user by giving detailed instructions how to assemble the theater. Its performance is based on advanced user-centered planning capabilities including the generation, repair, and explanation of plans.

Theoretical results on HTN planning are mostly related to the plan existence problem. In this paper, we study the structure of the generated plans in terms of the language they produce. We show that such languages are always context-sensitive. Furthermore we identify certain subclasses of HTN planning problems which generate either regular or context-free languages. Most importantly we have discovered that HTN planning problems, where preconditions and effects are omitted, constitute a new class of languages that lies strictly between the context-free and context-sensitive languages.