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.

Domain model acquisition is the problem of learning the structure of a state-transition system from some input data, typically example transition sequences. Recent work has shown that it is possible to learn action costs of PDDL models, given the overall costs of individual plans. In this work we have explored the extension of these ideas to narrative planning where cost can represent a variety of features (e.g. tension or relationship strength) and where exact solutions don’t exist. Hence in this paper we generalise earlier results to show that when an exact solution does not exist, a best-fit costing can be generated. This approach is of particular interest in the context of plan-based narrative generation where the input cost functions are based on subjective input. In order to demonstrate the effectiveness of the approach, we have learnt models of narratives using subjective measures of narrative tension. These were obtained with narratives (presented as video in this case) that were encoded as action traces, and then scored for subjective narrative tension by viewers. This provided a collection of input plan traces for our domain model acquisition system to learn a best-fit model. Using this learnt model we demonstrate how it can be used to generate new narratives that fit different target levels of tension.

This paper addresses the challenge of automated numeric domain model acquisition from observations. Many industrial and commercial applications of planning technology rely on numeric planning models. For example, in the area of autonomous systems and robotics, an autonomous robot often has to reason about its position in space, power levels and storage capacities. It is essential for these models to be easy to construct. Ideally, they should be automatically constructed. Learning the structure of planning domains from observations of action traces has produced successful results in classical planning. In this work, we present the first results in generalising approaches from classical planning to numeric planning. We restrict the numeric domains to those that include fixed action costs. Taking the finite state automata generated by the LOCM family of algorithms, we learn costs associated with machines; specifically to the object transitions and the state parameters. We learn action costs from action traces (with only the final cost of the plans as extra information) using a constraint programming approach. We demonstrate the effectiveness of this approach on standard benchmarks.

The evolution in precision manufacturing has resulted in the requirement to produce and maintain more accurate machine tools. This new requirement coupled with desire to reduce machine tool downtime places emphasis on the calibration procedure during which the machine’s capabilities are assessed. Machine tool downtime is significant for manufacturers because the machine will be unavailable for manufacturing use, therefore wasting the manufacturer’s time and potentially increasing lead-times for clients. In addition to machine tool downtime, the uncertainty of measurement, due to the schedule of the calibration plan, has significant implications on tolerance conformance, resulting in an increased possibility of false acceptance and rejection of machined parts. The work presented in this paper is focussed on expanding a developed temporal optimisation model to reduce the uncertainty of measurement. Encoding the knowledge in regular PDDL requires the discretization of non-linear, continuous temperature change and implementing the square root function. The implementation shows that not only can domainindependent automated planning reduce machine downtime by 10.6% and the uncertainty of measurement by 59%, it is also possible to optimise both metrics reaching a compromise that is on average 9% worse that the best-known solution for each individual metric.

Considerable effort has been spent extending the scope of planning beyond propositional domains to include, for example, time and numbers. Each extension has been designed as a separate specific semantic enrichment of the underlying planning model, with its own syntax and customised integration into a planning algorithm. Inspired by work on SAT Modulo Theories (SMT) in the SAT community, we develop a modelling language and planner that treat arbitrary first order theories as parameters. We call the approach Planning Modulo Theories (PMT). We introduce a modular language to represent PMT problems and demonstrate its benefits over PDDL in expressivity and compactness. We present a generalisation of the $h_{max}$ heuristic that allows our planner, PMTPlan, to automatically reason about arbitrary theories added as modules. Over several new and existing benchmarks, exploiting different theories, we show that PMTPlan can significantly out-perform an existing planner using PDDL models.

Partial Refinement A-Star (PRA* is an abstraction technique, based on clustering nearby nodes in graphs, useful in large path-planning problems. Abstracting the underlying graph yields a simpler problem whose solution can be used, by refinement, as a guide to a solution to the original problem. A fruitful way to view domain independent planning problems is as a collection of multi-valued variables that must perform synchronised transitions through graphs of possible values, where the edges are defined by the domain actions. Planning involves finding efficient paths through Domain Transition Graphs (DTGs). In problems where these graphs are large, planning can be prohibitively expensive. In this paper we explore two ways to exploit PRA* in DTGs.

One approach to the problem of formulating domain models for planning is to learn the models from example action sequences. The LOCM system demonstrated the feasibility of learning domain models from example action sequences only, with no observation of states before, during or after the plans. LOCM uses an object-centred representation, in which each object is represented by a single parameterised state machine. This makes it powerful for learning domains which fit within that representation, but there are some well-known domains which do not. This paper introduces LOCM2, a novel algorithm in which the domain representation of LOCM is generalised to allow multiple parameterised state machines to represent a single object. This extends the coverage of domains for which an adequate domain model can be learned. The LOCM2 algorithm is described and evaluated by testing domain learning from example plans from published results of past International Planning Competitions.