SHOP2: An HTN Planning System

Efficiency-Driven Adaptive Task Planning for Household Robot Based on Hierarchical Item-Environment Cognition

Task planning focused on household robots represents a conventional yet complex research domain, necessitating the development of task plans that enable robots to execute unfamiliar household services. This area has garnered significant research interest due to its extensive applications in robotics, particularly concerning household robots. Nevertheless, the majority of task planning methodologies exhibit suboptimal performance regarding the success and efficiency of completing household tasks, primarily due to a lack of cognitive capacity of household items and home environments. To address these challenges, we propose an efficiency-driven adaptive task planning approach based on hierarchical item-environment cognition. Initially, we establish a multiple semantic attribute-based priori knowledge (MSAPK) framework to facilitate the attributive representation of household items. Utilizing MSAPK, we develop a long short-term memory (LSTM) based item cognition model that assigns relevant attributes and substitutes to specified household items, thereby enhancing the cognitive capabilities of household robots at the attribute level. Subsequently, we construct an environment cognition model that delineates the relationships between household items and room types, enabling household robots to locate target items more efficiently. Through hierarchical item-environment cognition, we introduce a strategy for adaptive task planning, empowering household robots to execute household tasks with both flexibility and efficiency. The generated plans are evaluated in both virtual and real-world experiments, with promising results affirming the effectiveness of our proposed methodology.

Comparing Plan Recognition Algorithms Through Standard Plan Libraries

Plan recognition deals with reasoning about the goals and execution process of an actor, given observations of its actions. It is one of the fundamental problems of AI, applicable to many domains, from user interfaces to cyber-security. Despite the prevalence of these approaches, they lack a standard representation, and have not been compared using a common testbed. This paper provides a first step towards bridging this gap by providing a standard plan library representation that can be used by hierarchical, discrete-space plan recognition and evaluation criteria to consider when comparing plan recognition algorithms. This representation is comprehensive enough to describe a variety of known plan recognition problems and can be easily used by existing algorithms in this class. We use this common representation to thoroughly compare two known approaches, represented by two algorithms, SBR and Probabilistic Hostile Agent Task Tracker (PHATT). We provide meaningful insights about the differences and abilities of these algorithms, and evaluate these insights both theoretically and empirically. We show a tradeoff between expressiveness and efficiency: SBR is usually superior to PHATT in terms of computation time and space, but at the expense of functionality and representational compactness. We also show how different properties of the plan library affect the complexity of the recognition process, regardless of the concrete algorithm used. Lastly, we show how these insights can be used to form a new algorithm that outperforms existing approaches both in terms of expressiveness and efficiency.

A Robotic Cognitive Control Framework for Collaborative Task Execution and Learning

In social and service robotics, complex collaborative tasks are expected to be executed while interacting with humans in a natural and fluent manner. In this scenario, the robotic system is typically provided with structured tasks to be accomplished, but must also continuously adapt to human activities, commands, and interventions. We propose to tackle these issues by exploiting the concept of cognitive control, introduced in cognitive psychology and neuroscience to describe the executive mechanisms needed to support adaptive responses and complex goal-directed behaviors. Specifically, we rely on a supervisory attentional system to orchestrate the execution of hierarchically organized robotic behaviors. This paradigm seems particularly effective not only for flexible plan execution but also for human-robot interaction, because it directly provides attention mechanisms considered as pivotal for implicit, non-verbal human-human communication. Following this approach, we are currently developing a robotic cognitive control framework enabling collaborative task execution and incremental task learning. In this paper, we provide a uniform overview of the framework illustrating its main features and discussing the potential of the supervisory attentional system paradigm in different scenarios where humans and robots have to collaborate for learning and executing everyday activities.

Few-Shot Induction of Generalized Logical Concepts via Human Guidance

We consider the problem of learning generalized first-order representations of concepts from a small number of examples. We augment an inductive logic programming learner with 2 novel contributions. First, we define a distance measure between candidate concept representations that improves the efficiency of search for target concept and generalization. Second, we leverage richer human inputs in the form of advice to improve the sample efficiency of learning. We prove that the proposed distance measure is semantically valid and use that to derive a PAC bound. Our experiments on diverse learning tasks demonstrate both the effectiveness and efficiency of our approach.

A Framework for User Adaptation and Profiling for Social Robotics in Rehabilitation

Physical rehabilitation therapies for children present a challenge, and its success-the improvement of the patient's condition-depends on many factors, such as the patient's attitude and motivation, the correct execution of the exercises prescribed by the specialist or his progressive recovery during the therapy. With the aim to increase the benefits of these therapies, social humanoid robots with a friendly aspect represent a promising tool not only to boost the interaction with the pediatric patient, but also to assist physicians in their work. To achieve both goals, it is essential to monitor in detail the patient's condition, trying to generate user profile models which enhance the feedback with both the system and the specialist. This paper describes how the project NAOTherapist-a robotic architecture for rehabilitation with social robots-has been upgraded in order to include a monitoring system able to generate user profile models through the interaction with the patient, performing user-adapted therapies. Furthermore, the system has been improved by integrating a machine learning algorithm which recognizes the pose adopted by the patient and by adding a clinical reports generation system based on the QUEST metric.

An Abstract Framework for Non-Cooperative Multi-Agent Planning

In non-cooperative multi-agent planning environments, it is essential to have a system that enables the agents’ strategic behavior. It is also important to consider all planning phases, i.e., goal allocation, strategic planning, and plan execution, in order to solve a complete problem. Currently, we have no evidence of the existence of any framework that brings together all these phases for non-cooperative multi-agent planning environments. In this work, an exhaustive study is made to identify existing approaches for the different phases as well as frameworks and different applicable techniques in each phase. Thus, an abstract framework that covers all the necessary phases to solve these types of problems is proposed. In addition, we provide a concrete instantiation of the abstract framework using different techniques to promote all the advantages that the framework can offer. A case study is also carried out to show an illustrative example of how to solve a non-cooperative multi-agent planning problem with the presented framework. This work aims to establish a base on which to implement all the necessary phases using the appropriate technologies in each of them and to solve complex problems in different domains of application for non-cooperative multi-agent planning settings.

Robot Assistance in Dynamic Smart Environments-A Hierarchical Continual Planning in the Now Framework

By coupling a robot to a smart environment, the robot can sense state beyond the perception range of its onboard sensors and gain greater actuation capabilities. Nevertheless, incorporating the states and actions of Internet of Things (IoT) devices into the robot’s onboard planner increases the computational load, and thus can delay the execution of a task. Moreover, tasks may be frequently replanned due to the unanticipated actions of humans. Our framework aims to mitigate these inadequacies. In this paper, we propose a continual planning framework, which incorporates the sensing and actuation capabilities of IoT devices into a robot’s state estimation, task planing and task execution. The robot’s onboard task planner queries a cloud-based framework for actuators, capable of the actions the robot cannot execute. Once generated, the plan is sent to the cloud back-end, which will inform the robot if any IoT device reports a state change affecting its plan. Moreover, a Hierarchical Continual Planning in the Now approach was developed in which tasks are split-up into subtasks. To delay the planning of actions that will not be promptly executed, and thus to reduce the frequency of replanning, the first subtask is planned and executed before the subsequent subtask is. Only information relevant to the current (sub)task is provided to the task planner. We apply our framework to a smart home and office scenario in which the robot is tasked with carrying out a human’s requests. A prototype implementation in a smart home, and simulator-based evaluation results, are presented to demonstrate the effectiveness of our framework.

A review of generalized planning

Generalized planningstudies the representation, computation and evaluation of solutions that are valid for multiple planning instances. These are topics studied since the early days of AI. However, in recent years, we are experiencing the appearance of novel formalisms to compactly represent generalized planning tasks, the solutions to these tasks (calledgeneralized plans) and efficient algorithms to compute generalized plans. The paper reviews recent advances in generalized planning and relates them to existing planning formalisms, such asplanning with domain control knowledgeand approaches forplanning under uncertainty, that also aim at generality.

Developing a Robot-Guided Interactive Simon Game for Physical and Cognitive Training

Enveloping cognitive or physical rehabilitation into a game highly increases the patients’ commitment with their treatment. Specially with children, keeping them motivated is a very time-consuming work, so therapists are demanding tools to help them with this task. NAOTherapist is a generic robotic architecture that uses Automated Planning techniques to autonomously drive noncontact upper-limb rehabilitation sessions for children with a humanoid NAO robot. Our aim is to develop more robotic games for this platform to enrich its variability and possibilities of interaction. The goal of this work is to present our first attempt to develop a different, more complex game that reuses the previous architecture. We contribute with the design description of a novel robotic Simon game that employs upper-limb poses instead of colors and could qualify as a cognitive and physical training. Statistics of evaluation tests with 14 adults and 56 children are displayed and the outcomes are analyzed in terms of human–robot interaction (HRI) quality. The results demonstrate the application-domain generalization capabilities of the NAOTherapist architecture and give an insight to further analyze the therapeutic benefits of the new developed Simon game.

Towards life-long adaptive agents: using metareasoning for combining knowledge-based planning with situated learning

We consider task planning for long-living intelligent agents situated in dynamic environments. Specifically, we address the problem of incomplete knowledge of the world due to the addition of new objects with unknown action models. We propose a multilayered agent architecture that uses meta-reasoning to control hierarchical task planning and situated learning, monitor expectations generated by a plan against world observations, forms goals and rewards for the situated reinforcement learner, and learns the missing planning knowledge relevant to the new objects. We use occupancy grids as a low-level representation for the high-level expectations to capture changes in the physical world due to the additional objects, and provide a similarity method for detecting discrepancies between the expectations and the observations at run-time; the meta-reasoner uses these discrepancies to formulate goals and rewards for the learner, and the learned policies are added to the hierarchical task network plan library for future re-use. We describe our experiments in the Minecraft and Gazebo microworlds to demonstrate the efficacy of the architecture and the technique for learning. We test our approach against an ablated reinforcement learning (RL) version, and our results indicate this form of expectation enhances the learning curve for RL while being more generic than propositional representations.

Combining a hierarchical task network planner with a constraint satisfaction solver for assembly operations involving routing problems in a multi-robot context

This work addresses the combination of a symbolic hierarchical task network planner and a constraint satisfaction solver for the vehicle routing problem in a multi-robot context for structure assembly operations. Each planner has its own problem domain and search space, and the article describes how both planners interact in a loop sharing information in order to improve the cost of the solutions. The vehicle routing problem solver gives an initial assignment of parts to robots, making the distribution based on the distance among parts and robots, trying also to maximize the parallelism of the future assembly operations evaluating during the process the dependencies among the parts assigned to each robot. Then, the hierarchical task network planner computes a scheduling for the given assignment and estimates the cost in terms of time spent on the structure assembly. This cost value is then given back to the vehicle routing problem solver as feedback to compute a better assignment, closing the loop and repeating again the whole process. This interaction scheme has been tested with different constraint satisfaction solvers for the vehicle routing problem. The article presents simulation results in a scenario with a team of aerial robots assembling a structure, comparing the results obtained with different configurations of the vehicle routing problem solver and showing the suitability of using this approach.

Automated Planning Enables Complex Protocols on Liquid-Handling Robots

Robotic automation in synthetic biology is especially relevant for liquid handling to facilitate complex experiments. However, research tasks that are not highly standardized are still rarely automated in practice. Two main reasons for this are the substantial investments required to translate molecular biological protocols into robot programs, and the fact that the resulting programs are often too specific to be easily reused and shared. Recent developments of standardized protocols and dedicated programming languages for liquid-handling operations addressed some aspects of ease-of-use and portability of protocols. However, either they focus on simplicity, at the expense of enabling complex protocols, or they entail detailed programming, with corresponding skills and efforts required from the users. To reconcile these trade-offs, we developed Roboliq, a software system that uses artificial intelligence (AI) methods to integrate (i) generic formal, yet intuitive, protocol descriptions, (ii) complete, but usually hidden, programming capabilities, and (iii) user-system interactions to automatically generate executable, optimized robot programs. Roboliq also enables high-level specifications of complex tasks with conditional execution. To demonstrate the system's benefits for experiments that are difficult to perform manually because of their complexity, duration, or time-critical nature, we present three proof-of-principle applications for the reproducible, quantitative characterization of GFP variants.

Grounding of HTN Planning Domain

Many Artificial Intelligence techniques have been developed for intelligent and autonomous systems to act and make rational decisions based on perceptions of the world state. Among these techniques, HTN (Hierarchical Task Network) planning is one of the most used in practice. HTN planning is based on expressive languages allowing to specify complex expert knowledge for real world domains. At the same time, many preprocessing techniques for classical planning were proposed to speed up the search. One of these technique, named grounding, consists in enumerating and instantiating all the possible actions from the planning problem descriptions. This technique has proven its effectiveness. Therefore, combining the expressiveness of HTN planning with the efficiency of the grounding preprocessing techniques used in classical planning is a very challenging issue. In this paper, we propose a generic algorithm to ground the domain representation for HTN planning. We show experimentally that grounding process improves the performances of state of the art HTN planners on a range of planning problems from the International Planning Competition (IPC).

A Method for Unifying the Representations of Domain Knowledge and Planning Algorithm in Hierarchical Task Network

Domain knowledge of hierarchical task network (HTN) usually involves logical expressions with predicates. One needs to master two different languages which are used to describe domain knowledge and implement planner. This has presented enormous challenges for most programmers who are not familiar with logical expressions. To solve the problem a method of state variable representation from the programmer’s point of view is introduced. This method has powerful expressivity and can unify the representations of domain knowledge and planning algorithm. In Pyhop a HTN planner written in Python, methods and operators are all as ordinary Python functions rather than using a special-purpose language. Pyhop uses a Python object that contains variable bindings and does not include a horn-clause inference engine for evaluating preconditions of operators and methods. By taking a simple travel-planning problem, it shows that the method is easy to understand and integrate planning into ordinary programming.

A Multi-Agent Extension of a Hierarchical Task Network Planning Formalism

Describing planning domains using a common formalism promotes greater reuse of research, allowing a fairer comparison between different planners and approaches. Common planning formalisms for single-agent planning are already well established (e.g., PDDL, STRIPS, and HTN), but currently there is a shortage of multi-agent planning formalisms with clear semantics. In this paper, we propose a multi-agent extension of the Hierarchical Task Network (HTN) planning formalism for multi-agent planning problems. Our formalism, the Multi-Agent Hierarchical Task Network (MA-HTN), can be used to specify and represent multi-agent planning domains and problems. We provide a grammar with semantics for the domain and problem representation, and describe two case studies with the translation from multi-agent systems developed in JaCaMo to our MA-HTN formalism.

Intelligent Process Adaptation in the SmartPM System

The increasing application of process-oriented approaches in new challenging dynamic domains beyond business computing (e.g., healthcare, emergency management, factories of the future, home automation, etc.) has led to reconsider the level of flexibility and support required to manage complex knowledge-intensive processes in such domains. A knowledge-intensive process is influenced by user decision making and coupled with contextual data and knowledge production, and involves performing complex tasks in the “physical” real world to achieve a common goal. The physical world, however, is not entirely predictable, and knowledge-intensive processes must be robust to unexpected conditions and adaptable to unanticipated exceptions, recognizing that in real-world environments it is not adequate to assume that all possible recovery activities can be predefined for dealing with the exceptions that can ensue. To tackle this issue, in this paper we present SmartPM, a model and a prototype Process Management System featuring a set of techniques providing support for automated adaptation of knowledge-intensive processes at runtime. Such techniques are able to automatically adapt process instances when unanticipated exceptions occur, without explicitly defining policies to recover from exceptions and without the intervention of domain experts at runtime, aiming at reducing error-prone and costly manual ad-hoc changes, and thus at relieving users from complex adaptations tasks. To accomplish this, we make use of well-established techniques and frameworks from Artificial Intelligence, such as situation calculus, IndiGolog and classical planning. The approach, which is backed by a formal model, has been implemented and validated with a case study based on real knowledge-intensive processes coming from an emergency management domain.