This paper appears in Twelfth International Workshop on Non-Monotonic Reasoning (NMR08). An earlier version of this paper appears in Fourth Multidisciplinary Workshop on Advances in Preference Handling (M-Pref08) at AAAI-08
In this paper, we address the problem of generating preferred plans by combining the procedural control knowledge specified by Hierarchical Task Networks (HTNs) with rich qualitative user preferences. The outcome of our work is a language for specifyin user preferences, tailored to HTN planning, together with a provably optimal preference-based planner, HTNPLAN, that is implemented as an extension of SHOP2. To compute preferred plans, we propose an approach based on forward-chaining heuristic search. Our heuristic uses an admissible evaluation function measuring the satisfaction of preferences over partial plans. Our empirical evaluation demonstrates the effectiveness of our HTNPLAN heuristics. We prove our approach sound and optimal with respect to the plans it generates by appealing to a situation calculus semantics of our preference language and of HTN planning. While our implementation builds on SHOP2, the language and techniques proposed here are relevant to a broad range of HTN planners.
Sun Yat-sen University（中山大学）；IRIT（图卢兹计算机科学研究所）；University of Toulouse（图卢兹大学）
Hierarchical Task Network (HTN) planning is showing its power in real-world planning. Although domain experts have partial hierarchical domain knowledge, it is time-consuming to specify all HTN methods, leaving them incomplete. On the other hand, traditional HTN learning approaches focus only on declarative goals, omitting the hierarchical domain knowledge. In this paper, we propose a novel learning framework to refine HTN methods via task insertion with completely preserving the original methods. As it is difficult to identify incomplete methods without designating declarative goals for compo…
Hierarchical Task Network (HTN) planning is a practical and efficient approach to planning when the 'standard operating procedures' for a domain are available. Like Belief-Desire-Intention (BDI) agent reasoning, HTN planning performs hierarchical and context-based refinement of goals into subgoals and basic actions. However, while HTN planners 'lookahead' over the consequences of choosing one refinement over another, BDI agents interleave refinement with acting. There has been renewed interest in making HTN planners behave more like BDI agent systems, e.g. to have a unified representation for …
In this short paper, we outline nine classical benchmarks submitted to the first hierarchical planning track of the International Planning competition in 2020. All of these benchmarks are based on the HDDL language. The choice of the benchmarks was based on a questionnaire sent to the HTN community. They are the following: Barman, Childsnack, Rover, Satellite, Blocksworld, Depots, Gripper, and Hiking. In the rest of the paper we give a short description of these benchmarks. All are totally ordered.
University of Maryland（马里兰大学）；Institute for Systems Research（系统研究学院）；IBM Watson Research Center（IBM沃森研究中心）
The SHOP2 planning system received one of the awards for distinguished performance in the 2002 International Planning Competition. This paper describes the features of SHOP2 which enabled it to excel in the competition, especially those aspects of SHOP2 that deal with temporal and metric planning domains.
Lehigh University（理海大学）；Wright State University（莱特州立大学）
The ability of an agent to change its objectives in response to unexpected events is desirable in dynamic environments. In order to provide this capability to hierarchical task network (HTN) planning, we propose an extension of the paradigm called task modifiers, which are functions that receive a task list and a state and produce a new task list. We focus on a particular type of problems in which planning and execution are interleaved and the ability to handle exogenous events is crucial. To determine the efficacy of this approach, we evaluate the performance of our task modifier implementati…
Many planning techniques have been developed to allow autonomous systems to act and make decisions based on their perceptions of the environment. Among these techniques, HTN ({\it Hierarchical Task Network}) planning is one of the most used in practice. Unlike classical approaches of planning. HTN operates by decomposing task into sub-tasks until each of these sub-tasks can be achieved an action. This hierarchical representation provide a richer representation of planning problems and allows to better guide the plan search and provides more knowledge to the underlying algorithms. In this paper…
The Hierarchical Task Network (HTN) formalism is used to express a wide variety of planning problems in terms of decompositions of tasks into subtaks. Many techniques have been proposed to solve such hierarchical planning problems. A particular technique is to encode hierarchical planning problems as classical STRIPS planning problems. One advantage of this technique is to benefit directly from the constant improvements made by STRIPS planners. However, there are still few effective and expressive encodings. In this paper, we present a new HTN to STRIPS encoding allowing to generate concurrent…
In this paper we present an interface between a symbolic planner and a geometric task planner, which is different to a standard trajectory planner in that the former is able to perform geometric reasoning on abstract entities---tasks. We believe that this approach facilitates a more principled interface to symbolic planning, while also leaving more room for the geometric planner to make independent decisions. We show how the two planners could be interfaced, and how their planning and backtracking could be interleaved. We also provide insights for a methodology for using the combined system, a…
Institute for Architecture of Application Systems（应用系统架构研究所）；University of Stuttgart（斯图加特大学）
Cloud providers are facing a complex problem in configuring software applications ready for deployment on their infrastructures. Hierarchical Task Network (HTN) planning can provide effective means to solve such deployment problems. We present an HTN planning domain that models deployment problems as found in realistic Cloud environments.
University Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；ONERA（法国航空航天实验室）
Real world applications of planning, like in industry and robotics, require modelling rich and diverse scenarios. Their resolution usually requires coordinated and concurrent action executions. In several cases, such planning problems are naturally decomposed in a hierarchical way and expressed by a Hierarchical Task Network (HTN) formalism. The PDDL language used to specify planning domains has evolved to cover the different planning paradigms. However, formulating real and complex scenarios where numerical and temporal constraints concur in defining a solution is still a challenge. Our propo…
The Hierarchical Task Network ({\sf HTN}) formalism is very expressive and used to express a wide variety of planning problems. In contrast to the classical {\sf STRIPS} formalism in which only the action model needs to be specified, the {\sf HTN} formalism requires to specify, in addition, the tasks of the problem and their decomposition into subtasks, called {\sf HTN} methods. For this reason, hand-encoding {\sf HTN} problems is considered more difficult and more error-prone by experts than classical planning problem. To tackle this problem, we propose a new approach (HierAMLSI) based on gra…
Johann Bernoulli Institute for Mathematics and Computer Science（约翰·伯努利数学与计算机科学研究所）；University of Groningen（格罗宁根大学）
Hierarchies are the most common structure used to understand the world better. In galaxies, for instance, multiple-star systems are organised in a hierarchical system. Then, governmental and company organisations are structured using a hierarchy, while the Internet, which is used on a daily basis, has a space of domain names arranged hierarchically. Since Artificial Intelligence (AI) planning portrays information about the world and reasons to solve some of world's problems, Hierarchical Task Network (HTN) planning has been introduced almost 40 years ago to represent and deal with hierarchies.…
Institute for Systems Research, Univ. of Maryland（马里兰大学系统研究学院）；Navy Center for Applied Research in AI, Naval Research Laboratory（美国海军研究实验室海军人工智能应用研究中心）
Hierarchical Task Network Planning with LLM-Generated Heuristics
University of Aberdeen, UK（爱丁堡大学（英国））；PUCRS, Brazil（巴西普埃布拉联邦大学）；University of Oxford, UK（牛津大学（英国））；Saarland University, Germany（萨尔大学（德国））；Universidade Federal do Rio Grande do Sul, Brazil（巴西里约格兰德 do 南大学）
Johann Bernoulli Institute for Mathematics and Computer Science（约翰·伯努利数学与计算机科学研究所）；University of Groningen（格罗宁根大学）
Hierarchical Task Network (HTN) planning uses task decomposition to plan for an executable sequence of actions as a solution to a problem. In order to reason effectively, an HTN planner needs expressive domain knowledge. For instance, a simplified HTN planning system such as JSHOP2 uses such expressivity and avoids some task interactions due to the increased complexity of the planning process. We address the possibility of simplifying the domain representation needed for an HTN planner to find good solutions, especially in real-world domains describing home and building automation environments…
CNRS（法国国家科学研究中心）；LAAS（系统分析与体系结构实验室）；Univ de Toulouse（图卢兹大学）；INSA（国立应用科学学院）
Hierarchical Task Network (HTN) planning is a popular approach that cuts down on the classical planning search space by relying on a given hierarchical library of domain control knowledge. This provides an intuitive methodology for specifying high-level instructions on how robots and agents should perform tasks, while also giving the planner enough flexibility to choose the lower-level steps and their ordering. In this paper we present the HATP (Hierarchical Agent-based Task Planner) planning framework which extends the traditional HTN planning domain representation and semantics by making the…
Carnegie Mellon University（卡内基梅隆大学）；Arizona State University（亚利桑那州立大学）
We propose automatically learning probabilistic Hierarchical Task Networks (pHTNs) in order to capture a user's preferences on plans, by observing only the user's behavior. HTNs are a common choice of representation for a variety of purposes in planning, including work on learning in planning. Our contributions are (a) learning structure and (b) representing preferences. In contrast, prior work employing HTNs considers learning method preconditions (instead of structure) and representing domain physics or search control knowledge (rather than preferences). Initially we will assume that the obs…
University of Texas, Dallas（得克萨斯大学达拉斯分校）；Indiana University, Bloomington（印第安纳大学伯明顿分校）；Washington State University（华盛顿州立大学）；University of Pennsylvania（宾夕法尼亚大学）
Planning with preferences has been employed extensively to quickly generate high-quality plans. However, it may be difficult for the human expert to supply this information without knowledge of the reasoning employed by the planner and the distribution of planning problems. We consider the problem of actively eliciting preferences from a human expert during the planning process. Specifically, we study this problem in the context of the Hierarchical Task Network (HTN) planning framework as it allows easy interaction with the human. Our experimental results on several diverse planning domains sh…
Optimizing delivery routes for last-mile logistics service is challenging and has attracted the attention of many researchers. These problems are usually modeled and solved as variants of vehicle routing problems (VRPs) with challenging real-world constraints (e.g., time windows, precedence). However, despite many decades of solid research on solving these VRP instances, we still see significant gaps between optimized routes and the routes that are actually preferred by the practitioners. Most of these gaps are due to the difference between what's being optimized, and what the practitioners ac…
University Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；Ulm University（乌尔姆大学）；The MITRE Corporation（MITRE公司）
Over the last year, the amount of research in hierarchical planning has increased, leading to significant improvements in the performance of planners. However, the research is diverging and planners are somewhat hard to compare against each other. This is mostly caused by the fact that there is no standard set of benchmark domains, nor even a common description language for hierarchical planning problems. As a consequence, the available planners support a widely varying set of features and (almost) none of them can solve (or even parse) any problem developed for another planner. With this pape…
Knexus Research Corporation（Knexus研究公司）；MITRE（MITRE公司）；Naval Research Laboratory（海军研究实验室）
There is an impressive body of work on developing heuristics and other reasoning algorithms to guide search in optimal and anytime planning algorithms for classical planning. However, very little effort has been directed towards developing analogous techniques to guide search towards high-quality solutions in hierarchical planning formalisms like HTN planning, which allows using additional domain-specific procedural control knowledge. In lieu of such techniques, this control knowledge often needs to provide the necessary search guidance to the planning algorithm, which imposes a substantial bu…
We present a declarative language, PP, for the high-level specification of preferences between possible solutions (or trajectories) of a planning problem. This novel language allows users to elegantly express non-trivial, multi-dimensional preferences and priorities over such preferences. The semantics of PP allows the identification of most preferred trajectories for a given goal. We also provide an answer set programming implementation of planning problems with PP preferences.
Arizona State University（亚利桑那州立大学）；Palo Alto Research Center（帕洛阿尔托研究中心）；Università degli Studi di Brescia（布雷西亚大学）；Free University of Bozen-Bolzano（博尔扎诺自由大学）；IBM India Research Laboratory（IBM印度研究院）
Current work in planning with preferences assume that the user's preference models are completely specified and aim to search for a single solution plan. In many real-world planning scenarios, however, the user probably cannot provide any information about her desired plans, or in some cases can only express partial preferences. In such situations, the planner has to present not only one but a set of plans to the user, with the hope that some of them are similar to the plan she prefers. We first propose the usage of different measures to capture quality of plan sets that are suitable for such …
Linköping University（林雪平大学）；Universitat Pompeu Fabra（庞培法布拉大学）；Institució Catalana de Recerca i Estudis Avançats (ICREA)（加泰罗尼亚研究与高等研究院（ICREA））
Width-based planning methods deal with conjunctive goals by decomposing problems into subproblems of low width. Algorithms like SIW thus fail when the goal is not easily serializable in this way or when some of the subproblems have a high width. In this work, we address these limitations by using a simple but powerful language for expressing finer problem decompositions introduced recently by Bonet and Geffner, called policy sketches. A policy sketch over a set of Boolean and numerical features is a set of sketch rules that express how the values of these features are supposed to change. Like …
In most common settings of Markov Decision Process (MDP), an agent evaluate a policy based on expectation of (discounted) sum of rewards. However in many applications this criterion might not be suitable from two perspective: first, in risk aversion situation expectation of accumulated rewards is not robust enough, this is the case when distribution of accumulated reward is heavily skewed; another issue is that many applications naturally take several objective into consideration when evaluating a policy, for instance in autonomous driving an agent needs to balance speed and safety when choosi…
University of Oxford（牛津大学）；Stanford University（斯坦福大学）
An important use of machine learning is to learn what people value. What posts or photos should a user be shown? Which jobs or activities would a person find rewarding? In each case, observations of people's past choices can inform our inferences about their likes and preferences. If we assume that choices are approximately optimal according to some utility function, we can treat preference inference as Bayesian inverse planning. That is, given a prior on utility functions and some observed choices, we invert an optimal decision-making process to infer a posterior distribution on utility funct…
In this paper, we study planning in stochastic systems, modeled as Markov decision processes (MDPs), with preferences over temporally extended goals. Prior work on temporal planning with preferences assumes that the user preferences form a total order, meaning that every pair of outcomes are comparable with each other. In this work, we consider the case where the preferences over possible outcomes are a partial order rather than a total order. We first introduce a variant of deterministic finite automaton, referred to as a preference DFA, for specifying the user's preferences over temporally e…
Beyond Itinerary Planning-A Real-World Benchmark for Multi-Turn and Tool-Using Travel Tasks
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学光林人工智能学院）；National University of Singapore（新加坡国立大学）；Beihang University（北航）；AMAP, Alibaba Group（阿里云实验室）
We present a formal language for specifying qualitative preferences over temporal goals and a preference-based planning method in stochastic systems. Using automata-theoretic modeling, the proposed specification allows us to express preferences over different sets of outcomes, where each outcome describes a set of temporal sequences of subgoals. We define the value of preference satisfaction given a stochastic process over possible outcomes and develop an algorithm for time-constrained probabilistic planning in labeled Markov decision processes where an agent aims to maximally satisfy its pref…
Heriot-Watt University（赫瑞-瓦特大学）；University of Edinburgh（爱丁堡大学）
Human interaction relies on a wide range of signals, including non-verbal cues. In order to develop effective Explainable Planning (XAIP) agents it is important that we understand the range and utility of these communication channels. Our starting point is existing results from joint task interaction and their study in cognitive science. Our intention is that these lessons can inform the design of interaction agents -- including those using planning techniques -- whose behaviour is conditioned on the user's response, including affective measures of the user (i.e., explicitly incorporating the …
Future of Humanity Institute（人类未来研究所）；University of Oxford（牛津大学）；Vector Institute（向量研究所）；University of Toronto（多伦多大学）；Machine Intelligence Research Institute（机器智能研究所）
Inverse reinforcement learning (IRL) attempts to infer human rewards or preferences from observed behavior. Since human planning systematically deviates from rationality, several approaches have been tried to account for specific human shortcomings. However, the general problem of inferring the reward function of an agent of unknown rationality has received little attention. Unlike the well-known ambiguity problems in IRL, this one is practically relevant but cannot be resolved by observing the agent's policy in enough environments. This paper shows (1) that a No Free Lunch result implies it i…
Prior work on generating explanations in a planning and decision-making context has focused on providing the rationale behind an AI agent's decision making. While these methods provide the right explanations from the explainer's perspective, they fail to heed the cognitive requirement of understanding an explanation from the explainee's (the human's) perspective. In this work, we set out to address this issue by first considering the influence of information order in an explanation, or the progressiveness of explanations. Intuitively, progression builds later concepts on previous ones and is k…
