In this paper, we describe an approach for learning planning domain models directly from natural language (NL) descriptions of activity sequences. The modelling problem has been identified as a bottleneck for the widespread exploitation of various technologies in Artificial Intelligence, including automated planners. There have been great advances in modelling assisting and model generation tools, including a wide range of domain model acquisition tools. However, for modelling tools, there is the underlying assumption that the user can formulate the problem using some formal language. And even in the case of the domain model acquisition tools, there is still a requirement to specify input plans in an easily machine readable format. Providing this type of input is impractical for many potential users. This motivates us to generate planning domain models directly from NL descriptions, as this would provide an important step in extending the widespread adoption of planning techniques. We start from NL descriptions of actions and use NL analysis to construct structured representations, from which we construct formal representations of the action sequences. The generated action sequences provide the necessary structured input for inducing a PDDL domain, using domain model acquisition technology. In order to capture a concise planning model, we use an estimate of functional similarity, so sentences that describe similar behaviours are represented by the same planning operator. We validate our approach with a user study, where participants are tasked with describing the activities occurring in several videos. Then our system is used to learn planning domain models using the participants' NL input. We demonstrate that our approach is effective at learning models on these tasks.

Empirical performance models play an important role in the development of planning portfolios that make a per-domain or per-problem configuration of its search components. Even though such portfolios have shown their power when compared to other systems in current benchmarks, there is no clear evidence that they are capable to differentiate problems (instances) having similar input properties (in terms of objects, goals, etc.) but fairly different runtime for a given planner. In this paper we present a study of empirical performance models that are trained using problems having the same configuration, with the objective of guiding the models to recognize the underlying differences existing among homogeneous problems. In addition we propose a set of new features that boost the prediction capabilities under such scenarios. The results show that the learned models clearly performed over random classifiers, which reinforces the hypothesis that the selection of planners can be done on a per-instance basis when configuring a portfolio.

This paper is an experience report on the results of an industry-led collaborative project aimed at automating the control of traffic flow within a large city centre. A major focus of the automation was to deal with abnormal or unexpected events such as roadworks, road closures or excessive demand, resulting in periods of saturation of the network within some region of the city. We describe the resulting system which works by sourcing and semantically enriching urban traffic data, and uses the derived knowledge as input to an automated planning component to generate light signal control strategies in real time. This paper reports on the development surrounding the planning component, and in particular the engineering, configuration and validation issues that arose in the application. It discusses a range of lessons learned from the experience of deploying automated planning in the road transport area, under the direction of transport operators and technology developers.

A challenging Earth-observing satellite scheduling problem was recently studied in (Frank, Do and Tran 2016) for which the best resolution approach so far on the proposed benchmark is a time-indexed Mixed Integer Linear Program (MILP) formulation. This MILP formulation produces feasible solutions but is not able to prove optimality or to provide tight optimality gaps, making it difficult to assess the quality of existing solutions. In this paper, we first introduce an alternative disjunctive MILP formulation that manages to close more than half of the instances of the benchmark. This MILP formulation is then relaxed to provide good bounds on optimal values for the unsolved instances. We then propose a CP Optimizer model that consistently outperforms the original time-indexed MILP formulation, reducing the optimality gap by more than 4 times. This Constraint Programming (CP) formulation is very concise: we give its complete OPL implementation in the paper. Some improvements of this CP model are reported resulting in an approach that produces optimal or near-optimal solutions (optimality gap smaller than 1%) for about 80% of the instances. Unlike the MILP formulations, it is able to quickly produce good quality schedules and it is expected to be flexible enough to handle the changing requirements of the application.

In this work, we investigate a multi-period Home Health Care Scheduling Problem (HHCSP) under stochastic service and travel times. We first model the deterministic problem as an integer linear programming model that incorporates real-world requirements, such as time windows, continuity of care, workload fairness, inter-visit temporal dependencies. We then extend the model to cope with uncertainty in durations, by introducing chance constraints into the formulation. We propose efficient solution approaches, which provide quantifiable near-optimal solutions and further handle the uncertainties by employing a sampling-based strategy. We demonstrate the effectiveness of our proposed approaches on instances synthetically generated by real-world dataset for both deterministic and stochastic scenarios.

In cooperative multiagent planning, it can often be beneficial for an agent to make commitments about aspects of its behavior to others, allowing them in turn to plan their own behaviors without taking the agent's detailed behavior into account. Extending previous work in the Bayesian setting, we consider instead a worst-case setting in which the agent has a set of possible environments (MDPs) it could be in, and develop a commitment semantics that allows for probabilistic guarantees on the agent's behavior in any of the environments it could end up facing. Crucially, an agent receives observations (of reward and state transitions) that allow it to potentially eliminate possible environments and thus obtain higher utility by adapting its policy to the history of observations. We develop algorithms and provide theory and some preliminary empirical results showing that they ensure an agent meets its commitments with history-dependent policies while minimizing maximum regret over the possible environments.

When planning actions to take on behalf of its human operator, a robot might be uncertain about its operator's reward function. We address the problem of how the robot should formulate an (approximately) optimal query to pose to the operator, given how its uncertainty affects which policies it should plan to pursue. We explain how a robot whose queries ask the operator to choose the best from among k choices can, without loss of optimality, restrict consideration to choices only over alternative policies. Further, we present a method for constructing an approximately-optimal policy query that enjoys a performance bound, where the method need not enumerate all policies. Finally, because queries posed to the operator of a robotic system are often expressed in terms of preferences over trajectories rather than policies, we show how our constructed policy query can be projected into the space of trajectory queries. Our empirical results demonstrate that our projection technique can outperform other known techniques for choosing trajectory queries, particularly when the number of trajectories the operator is asked to compare is small.

We investigate solvable-unsolvable phase transitions in the single-machine scheduling (SMS) problem. SMS is at the core of practical problems such as telescope and satellite scheduling and manufacturing. To study the solvability phase transition, we construct a variety of instance families param-eterized by the set of the processing times, the window size (deadline minus release time), and the horizon. We empirically establish the phase transition and look for an easy-hard- easy pattern for this family using several common solvers. While in many combinatorial problems a phase transition coincides with typically hard instances, whether or not that is the case with SMS remains an open question, and merits further study.

Recent work in multi-agent path planning has provided a number of optimal and suboptimal solvers that can efficiently find solutions to problems of growing complexity. Solvers based on Conflict-Based Search (CBS) combine single-agent solvers with shared constraints between agents to find feasible solutions. Suboptimal variants of CBS introduce alternate heuristics to avoid conflicts. In this paper we study the multi-agent planning problem in the context of non-holonomic vehicles planning on a lattice. We propose that in addition to using heuristics to avoid conflicts, we can plan using a hierarchy of movement constraints to efficiently avoid conflicts. We develop a new extension to the CBS algorithm, CBS with constraint layering (CBS+CL), which iteratively applies different movement constraint models during the CBS planning process. Our results show that this approach allows us to solve for about 2.4 times more agents in the same amount of time when compared to regular CBS without using a constraint hierarchy.

For the past 25 years, heuristic search has been used to solve domain-independent probabilistic planning problems, but with heuristics that determinise the problem and ignore precious probabilistic information. To remedy this situation, we explore the use of occupation measures, which represent the expected number of times a given action will be executed in a given state of a policy. By relaxing the well-known linear program that computes them, we derive occupation measure heuristics -- the first admissible heuristics for stochastic shortest path problems (SSPs) taking probabilities into account. We show that these heuristics can also be obtained by extending recent operator-counting heuristic formulations used in deterministic planning. Since the heuristics are formulated as linear programs over occupation measures, they can easily be extended to more complex probabilistic planning models, such as constrained SSPs (C-SSPs). Moreover, their formulation can be tightly integrated into i-dual, a recent LP-based heuristic search algorithm for (constrained) SSPs, resulting in a novel probabilistic planning approach in which policy update and heuristic computation work in unison. Our experiments in several domains demonstrate the benefits of these new heuristics and approach.