A review and comparison of ontology-based approaches to robot autonomy

Abstract

Within the next decades, robots will need to be able to execute a large variety of tasks autonomously in a large variety of environments. To relax the resulting programming effort, a knowledge-enabled approach to robot programming can be adopted to organize information in re-usable knowledge pieces. However, for the ease of reuse, there needs to be an agreement on the meaning of terms. A common approach is to represent these terms using ontology languages that conceptualize the respective domain. In this work, we will review projects that use ontologies to support robot autonomy. We will systematically search for projects that fulfill a set of inclusion criteria and compare them with each other with respect to the scope of their ontology, what types of cognitive capabilities are supported by the use of ontologies, and which is their application domain.

Reliability and Safety of Autonomous Systems Based on Semantic Modelling for Self-Certification

“…They covered a spectrum of cognitive functions, which according to the classification made by [42,43] are: recognition and categorization, decision making and choice, perception and situation assessment, prediction and monitoring, problem solving and planning, reasoning and belief maintenance, execution and action, interaction and communication, and remembering, reflection, and learning. The ontology scope of these prior works varies and it depends on the functionalities of the target robotic system, i.e., concepts that were modelled in the ontology are related to: object names, environment, affordance, action and task, activity and behaviour, plan and method, capability and skill, hardware components, software components, interaction and communication [44,45].…”

“…The core ontology was utilised in some projects such as [50,51]. However, the core ontology is too generic for complex applications, overlooking complex mission and robotic platform sub-system interdependencies [44].…”

“…It has been used in several research initiatives including the European projects RoboHow [52], RoboEarth [46], SAPHARI [53], and SHERPA [54]. The current development of KNOWROB has shifted towards the integration of simulation and rendering techniques into a hybrid knowledge framework [44,45] The KNOWROB framework does not have any diagnosis or prognosis features.…”

“…When investigating the complexity of the ontology it has to be in bytes as both raw data and ontology at the bottom are measured in bytes. Table 2 compares (following an approach similar to [44,45]) different systems that utilized ontologies. It is not the theme of this paper to discuss the system integration or architectural aspects, but the focus here is on: (1) the semantic relationships and the related automatons that are used for modelling the hardware components and the links between them, (2) the semantic rules for the faults detection, diagnosis, and prognosis (FDDP) feature.…”

Conversational ontology operator: patient-centric vaccine dialogue management engine for spoken conversational agents

“…However, ontologies have also been known to provide support in automated planning – an AI sub-field for using model-based behavior methods for agents. For example, Olivares-Alarcos and colleagues recently reviewed and identified ontologies for mechatronic-related research [ 11 ]. Essentially, ontologies can provide software agents with intelligence and reasoning on how to respond in an environment with other virtual or physical agents, along with sharing an understanding of the environment among the agents.…”