Successfully accomplishing everyday manipulation tasks requires robots to have substantial knowledge about the objects they interact with, the environment they operate in as well as about the properties and effects of the actions they perform. Often, this knowledge is implicitly contained in manually written control programs, which makes it hard for the robot to adapt to newly acquired information or to re-use knowledge in a different context. By explicitly representing this knowledge, control decisions can be formulated as inference tasks which can be sent as queries to a knowledge base. This allows the robot to take all information it has at query time into account to generate answers, leading to better flexibility, adaptability to changing situations, robustness, and the ability to re-use knowledge once acquired. In this paper, we report on our work towards a practical and grounded knowledge representation and inference system. The system is specifically designed to meet the challenges created by using knowledge processing techniques on autonomous robots, including specialized inference methods, grounding of symbolic knowledge in the robot's control structures, and the acquisition of the different kinds of knowledge a robot needs.

In this paper we examine the case of a mobile robot that is part of a human-robot urban search and rescue (USAR) team. During USAR scenarios, we would like the robot to have a geometrical-functional understand- ing of space, using which it can infer where to perform planned tasks in a manner that mimics human behav- ior. We assess the situation awareness of rescue work- ers during a simulated USAR scenario and use this as an empirical basis to build our robot’s spatial model. Based upon this spatial model, we present “functional map- ping” as an approach to identify regions in the USAR environment where planned tasks are likely to be opti- mally achievable. The system is deployed and evaluated in a simulated rescue scenario.

This paper describes an attempt to provide more intelligence to industrial robotics and automation systems. We develop an architecture to integrate disparate knowledge representations used in different places in robotics and automation. This knowledge integration framework, a possibly distributed entity, abstracts the components used in design or production as data sources, and provides a uniform access to them via standard interfaces. Representation is based on the ontology formalizing the process, product and resource triangle, where skills are considered the common element of the three. Production knowledge is being collected now and a preliminary version of KIF undergoes verification.

This paper takes new look on language and knowledge modelling for corpus linguistics. Using ideas of Chaitin, a line of argument is made against language/knowledge separation in Natural Language Processing. A simplistic model, that generalises approaches to language and knowledge, is proposed. One of hypothetical consequences of this model is Strong AI.

In a world where communication and information sharing are at the heart of our business, the terminology needs are most pressing. It has become imperative to identify the terms used and defined in a consensual and coherent way while preserving linguistic diversity. To streamline and strengthen the process of acquisition, representation and exploitation of scenarii of train accidents, it is necessary to harmonize and standardize the terminology used by players in the security field. The research aims to significantly improve analytical activities and operations of the various safety studies, by tracking the error in system, hardware, software and human. This paper presents the contribution of ontology to modeling scenarii for rail accidents through a knowledge model based on a generic ontology and domain ontology. After a detailed presentation of the state of the art material, this article presents the first results of the developed model.

We propose a novel, type-elimination-based method for reasoning in the description logic SHIQbs including DL-safe rules. To this end, we first establish a knowledge compilation method converting the terminological part of an ALCIb knowledge base into an ordered binary decision diagram (OBDD) which represents a canonical model. This OBDD can in turn be transformed into disjunctive Datalog and merged with the assertional part of the knowledge base in order to perform combined reasoning. In order to leverage our technique for full SHIQbs, we provide a stepwise reduction from SHIQbs to ALCIb that preserves satisfiability and entailment of positive and negative ground facts. The proposed technique is shown to be worst case optimal w.r.t. combined and data complexity and easily admits extensions with ground conjunctive queries.

Social media analytics has been attracting considerable attention in both research and industry due to the increasing popularity of social media usage. As a subset, social media monitoring describes the process of continuous monitoring of a subject matter in social media. From our point of view, the key requirements for such systems are i) high throughput and real-time processing of incoming data, ii) a user-friendly way to define complex situations of interests that make use of formalized background knowledge and iii) capabilities to perform actions based on gained insights instead of a pure monitoring system. In this paper, we propose a system for (pro) active, real-time social media monitoring. Firstly, we describe the conceptual architecture of our system and necessary pre-processing steps. Secondly, we introduce our concept of semantic requests that is capable to extend event pattern definitions with background knowledge. Finally, we show the usefulness of this system in two different domains: Real-time political opinion tracking and proactive establishment of relationships with consumers in order to perform a new form of real-time marketing. The main advantage of our approach is a simplified, expressive way to formulate event patterns in social media applications.

Current techniques for query answering over DL-Lite ontologies have severe limitations in practice, since they either produce complex queries that are inefficient during execution, or require expensive data pre-processing. In light of this, we present two complementary sets of results that aim at improving the overall peformance of query answering systems. We show how to create ABox repositories that are complete w.r.t. a significant portion of DL-Lite TBoxes, but where the data is not explicitly expanded. Second, we show how to characterize ABox completeness by means of dependencies, and how to use these and equivalence to optimize DL-Lite TBoxes. These results allow us to reduce the cost of query rewriting, often dramatically, and to generate highly efficient queries. We have implemented a novel system for query answering over DL-Lite ontologies that incorporates these techniques, and we present a series of data-intensive evaluations that show their effectiveness.

We are witnessing an explosion of available data from the Web, government authorities, scientific databases, sensors and more. Such datasets could benefit from the introduction of rule sets encoding commonly accepted rules or facts, application- or domain-specific rules, commonsense knowledge etc. This raises the question of whether, how, and to what extent knowledge representation methods are capable of handling the vast amounts of data for these applications. In this paper, we consider nonmonotonic reasoning, which has traditionally focused on rich knowledge structures. In particular, we consider defeasible logic, and analyze how parallelization, using the MapReduce framework, can be used to reason with defeasible rules over huge data sets. Our experimental results demonstrate that defeasible reasoning with billions of data is performant, and has the potential to scale to trillions of facts.

This paper presents a characterisation of and definitions for the phenomenon of masking in the context of justifications for entailments in ontologies. In essence masking is present within a justification, over a set of justifications, or over a complete ontology when the number of justifications for an entailment does not reflect the number of reasons for that entailment. Four types of masking are defined in this paper: Internal Masking, Cross Masking, External Masking and Shared Core Masking. The results of an empirical study are presented which shows that the phenomenon of masking is prevalent throughout ontologies with non-trivial entailments in the NCBO BioPortal corpus. Out of 72 ontologies, 53 exhibited some form of masking, with 9 ontologies exhibiting internal masking, 23 ontologies exhibiting external masking, and 53 ontologies exhibiting shared core masking.