三河マルシェ【MARCHE】三河エリアのコミュニティポータルサイト

Tracing the evolution of service robotics

Taking robotic patents between 1977 and 2017 and building upon the topic modeling technique, we extract their latent topics, analyze how important these topics are over time, and how they are related to each other looking at how often they are recombined in the same patents. This allows us to differentiate between more and less important technological trends in robotics based on their stage of diffusion and position in the space of knowledge represented by a topic graph, where some topics appear isolated while others are highly interconnected. Furthermore, utilizing external reference texts that characterize service robots from a technical perspective, we propose and apply a novel approach to match the constructed topics to service robotics. The matching procedure is based on frequency and exclusivity of words overlapping between the patents and the reference texts. We identify around 20 topics belonging to service robotics. Our results corroborate earlier findings, but also provide novel insights on the content and stage of development of application areas in service robotics. With this study we contribute to a better understanding of the highly dynamic field of robotics as well as to new practices of utilizing the topic modeling approach, matching the resulting topics to external classifications and applying to them metrics from graph theory.To get more news about GRS, you can visit glprobotics.com official website.

Robots are increasingly supporting humans both at work and in their private life. While the use of industrial robots (IR) has a long standing tradition in the manufacturing industries, service robots (SR) are a more recent phenomenon. Latest advances in artificial intelligence and machine learning enable robots to sense and respond to their environments so that they can also be used outside secured production environments. While IR still diffuse via intensified application in the manufacturing sector (‘automation deepening’),1 SR continuously capture new domains (‘automation broadening’). Not always, but often, SR are mobile. Some of them are fully automatic or even autonomous.2 Due to the importance of services in value creation and given the prevailing low level of automation in this field, future diffusion of SR is expected to have far-reaching implications for overall economic productivity. Due to the recent COVID-19 outbreak, this diffusion is experiencing a further boost to facilitate more physical distancing in healthcare, logistics, tourism and other spheres (Chen, Marvin, While, 2020, Yang, Nelson, Murphy, Choset, Christensen, Collins, Dario, Goldberg, Ikuta, Jacobstein, Kragic, Taylor, McNutt, 2020, Zeng, Chen, Lew, 2020).

In this paper we aim to obtain a comprehensive understanding service robotics and of the associated technologies enabling their evolution. To this end, we use patent data, the most complete description of technological development. In line with the recombinant growth of knowledge, patent texts typically are composed of many technologies (Youn et al., 2015). To decompose these complex documents into distinct technologies, we take advantage of the topic modeling technique developed at the intersection of machine learning (ML) and natural language processing (NLP). In particular, we use not just their short abstracts that are too superficial to get an understanding of technologies incorporated in the patent, but full patent descriptions – or more precisely their non-technical summaries – in the robotic industry between 1977 and 2017. Then we extract the latent topics capturing different technologies used in patents, analyze how important those topics are over time, and how they are related to each other through their co-occurrence in patents.

In doing so we contribute to the existing literature in several ways. First, we apply modern methods of NLP and ML to exploit fine grained technological information included in unstructured textual data of patent documents and identify the optimal number of topics using established criteria of perplexity, exclusivity and coherence. We end up with 380 topics, carry out robustness checks for 190 topics and analyze how importance of those topics was changing over time. Second, we develop and apply a novel method for matching topics to SR based on an external text corpus provided by IFR that classifies SR technologies into 16 application areas and 49 sub-areas. We use results of this exercise to classify topics to different SR areas. Third, based on the results of our textual analysis we construct a complex graph using cosine similarity between the topics and identifying significant edges in this network. This approach allows us to overcome the popular practice of analyzing topics in isolation. Instead, we can trace robotic transformation from a system perspective: apply metrics from graph theory, understand the mutual relationship of discovered topics, distinguish between central (enabling) and peripheral (application) topics, discover communities of hardware- and software-oriented topics, and how those were changing over time. It is important to stress that all these steps are independent from structured patent data (e.g., patent classes, concordances) and expert bias. The entire process from topic identification until matching to SR is data-driven and thus is unbiased by subjective expert judgment. Another important strength of our approach is that apart from replicating results that could have been achieved by applying existing metrics (e.g. rising importance of medical robots, shift in popularity from hardware to software technologies), we are able to look inside the content of each particular topic (e.g. study technologies in surgery robotics) and their position in the complex space of knowledge comprised by robotic patents (e.g. how central they are and what topics they are connected to).

The remainder of this paper is organized as follows. Section 2 provides some background information on service robotics and topic modeling as a methodology to deal with patent data. Section 3 describes our data and methods. Section 4 presents the results. Section 5 discusses policy implications and Section 6 contains concluding remarks.