KnowRob: A knowledge processing infrastructure for cognition-enabled robots

Abstract

Autonomous service robots will have to understand vaguely described tasks, such as "set the table" or "clean up". Performing such tasks as intended requires robots to fully, precisely, and appropriately parameterize their low-level control programs. We propose knowledge processing as a computational resource for enabling robots to bridge the gap between vague task descriptions and the detailed information needed to actually perform those tasks in the intended way. In this article, we introduce the KNOWROB knowledge processing system that is specifically designed to provide autonomous robots with the knowledge needed for performing everyday manipulation tasks. The system allows the realization of "virtual knowledge bases": collections of knowledge pieces that are not explicitly represented but computed on demand from the robot's internal data structures, its perception system, or external sources of information. This article gives an overview of the different kinds of knowledge, the different inference mechanisms, and interfaces for acquiring knowledge from external sources, such as the robot's perception system, observations of human activities, Web sites on the Internet, as well as Web-based knowledge bases for information exchange between robots. We evaluate the system's scalability and present different integrated experiments that show its versatility and comprehensiveness.

Learning Actions to Improve the Perceptual Anchoring of Objects

“…Our motivation is to improve the learning of object labels, rather than the classification of specific human motions. As an alternative to learning complex actions (or activities), Tenorth and Beetz (2013) introduced a complementary knowledge processing system (KnowRob). The authors introduced the use of commonsense information on a larger scale (such as information from the Web) to reason upon a perceived scene and inferring possible actions (or action primitives) for robot manipulation tasks.…”

Towards knowledge modeling and manipulation technologies: A survey

“…Mostly, in a LRS, knowledge is acquired manually from domain experts through interviews, where experts communicate their knowledge using questionnaires (Connaghan et al, 2013;Dimitroula et al, 2001;Selva & Crawley, 2012). However, knowledge in forms of rules can be acquired automatically, such as RUBRIC which constructs rules from thesauri (Minkoo et al, 2000) and semi-automated like KnowRob, which automatically acquires information from different knowledge sources with the aid of human for correcting mistakes and aligning imported knowledge sources (Tenorth & Beetz, 2013).…”

“…Prolog is the programming language used mostly for knowledge representation in logical rule-based systems, as seen in WUENIC (Kowalski & Burton, 2012), sports coaching (Connaghan et al, 2013), KnowRob (Tenorth & Beetz, 2013) and online poker agent (Teofilo et al, 2014). Other development tools used are CLIPS, for the implementation of FUNAGES (Dimitroula et al, 2001).…”

Towards an Explanation Generation System for Robots: Analysis and Recommendations

“…For instance, the Oro framework creates a sophisticated ontology and uses it to justify observed inconsistencies on a robot [26]. The KnowRob architecture for service robots uses knowledge bases created from different sources to perform limited analysis of the reasons for unexpected observations [27]. Other examples include the use of ASP for planning and diagnostics by one or more simulated robot housekeepers [28] and mobile robot teams [29] and by robots reasoning about domain knowledge learned through natural language interactions with humans [30].…”