Towards an optimal design of natural human interaction mechanisms for a service robot with ancillary way-finding capabilities in industrial environments

Aceta, Cristina; Kildal, Johan; Fernández, Izaskun; Soroa, Aitor

doi:10.1080/21693277.2020.1870011

Cited by 2 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, training data is scarce in this domain, and the use of machine-learning approaches is still very limited. Although there are attempts to combine rules and machine-learning techniques [18,19], rule-based approaches are generally used in these scenarios due to their specific characteristics [5,11]. As a consequence, most task-oriented dialogue systems for industrial scenarios are heavily adapted to the task they have been designed for and cannot be reused in other contexts, and developing new ones for new use cases is bound to expert work and high time and costs [5].…”

Section: Related Workmentioning

confidence: 99%

“…The main function of the key element extraction component, thoroughly described in [18], is to obtain the relevant key elements from a transcribed user voice command that conveys a piece of information. Furthermore, it has been designed so as to learn over time from new interactions by generating training data semiautomatically to implement a supervised key element extraction component in the future.…”

Section: Key Element Extractionmentioning

confidence: 99%

“…Foma is a software that is able to convert regular expressions-defined in grammars-to finite-state automata, which is used in different natural language applications, such as morphological analysis. In this case, definitions and rules [18] need to be defined. On the one hand, definitions-most of which can be reused between use casesmodel the different syntactic structures that are of interest for the use case (e.g., a noun phrase can appear in the syntactic tree as a noun-container-or as a noun + adjective-blue container).…”

Section: Key Element Extractionmentioning

confidence: 99%

“…Once enough data has been obtained to train a supervised model, the Foma subcomponent is to be complemented with a machine-learning-based subcomponent that will tag the user command at word level, using BIO tags, considering the syntactic tree obtained from Freeling [18]. In this case, the interpreter will parse these word-BIO tag pairs to obtain the set of keywords for the command in the format described above.…”

Section: Key Element Extractionmentioning

confidence: 99%

See 3 more Smart Citations

KIDE4I: A Generic Semantics-Based Task-Oriented Dialogue System for Human-Machine Interaction in Industry 5.0

2022

Self Cite

View full text Add to dashboard Cite

In Industry 5.0, human workers and their wellbeing are placed at the centre of the production process. In this context, task-oriented dialogue systems allow workers to delegate simple tasks to industrial assets while working on other, more complex ones. The possibility of naturally interacting with these systems reduces the cognitive demand to use them and triggers acceptation. Most modern solutions, however, do not allow a natural communication, and modern techniques to obtain such systems require large amounts of data to be trained, which is scarce in these scenarios. To overcome these challenges, this paper presents KIDE4I (Knowledge-drIven Dialogue framEwork for Industry), a semantic-based task-oriented dialogue system framework for industry that allows workers to naturally interact with industrial systems, is easy to adapt to new scenarios and does not require great amounts of data to be constructed. This work also reports the process to adapt KIDE4I to new scenarios. To validate and evaluate KIDE4I, it has been adapted to four use cases that are relevant to industrial scenarios following the described methodology, and two of them have been evaluated through two user studies. The system has been considered as accurate, useful, efficient, not demanding cognitively, flexible and fast. Furthermore, subjects view the system as a tool to improve their productivity and security while carrying out their tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Key Element Extractionmentioning

confidence: 99%

Section: Key Element Extractionmentioning

confidence: 99%

Section: Key Element Extractionmentioning

confidence: 99%

See 2 more Smart Citations

KIDE4I: A Generic Semantics-Based Task-Oriented Dialogue System for Human-Machine Interaction in Industry 5.0

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although they are proven to be widely used, machine-learning-based methods require of great amounts of data to train the systems, which is not easy to obtain for industrial task-oriented dialogue systems, and rules are often generated for both natural language understanding and dialogue management. However, constructing rules is time and cost consuming and may be prone to errors, and supervised machine learning techniques are being added to the paradigm by combining them with these rules to optimize results [20,21].…”

Section: Related Workmentioning

confidence: 99%

TODO: A Core Ontology for Task-Oriented Dialogue Systems in Industry 4.0

Aceta¹,

Fernández²,

Soroa³

2021

Studies on the Semantic Web

Self Cite

View full text Add to dashboard Cite

Nowadays, the demand in industry of dialogue systems to be able to naturally communicate with industrial systems is increasing, as they allow to enhance productivity and security in these scenarios. However, adapting these systems to different use cases is a costly process, due to the complexity of the scenarios and the lack of available data. This work presents the Task-Oriented Dialogue management Ontology (TODO), which aims to provide a core and complete base for semantic-based task-oriented dialogue systems in the context of industrial scenarios in terms of, on the one hand, domain and dialogue modelling and, on the other hand, dialogue management and tracing support. Furthermore, its modular structure, besides grouping specific knowledge in independent components, allows to easily extend each of the modules, attending the necessities of the different use cases. These characteristics allow an easy adaptation of the ontology to different use cases, with a considerable reduction of time and costs. So as to demonstrate the capabilities of the the ontology by integrating it in a task-oriented dialogue system, TODO has been validated in real-world use cases. Finally, an evaluation is also presented, covering different relevant aspects of the ontology.

show abstract

Towards an optimal design of natural human interaction mechanisms for a service robot with ancillary way-finding capabilities in industrial environments

Abstract: Soroa (2021) Towards an optimal design of natural human interaction mechanisms for a service robot with ancillary wayfinding capabilities in industrial environments,

Cited by 2 publications

References 25 publications

KIDE4I: A Generic Semantics-Based Task-Oriented Dialogue System for Human-Machine Interaction in Industry 5.0

KIDE4I: A Generic Semantics-Based Task-Oriented Dialogue System for Human-Machine Interaction in Industry 5.0

TODO: A Core Ontology for Task-Oriented Dialogue Systems in Industry 4.0

Contact Info

Product

Resources

About