Are we prepared for the Rise of Service Robots? - A Review on Acceptance Measurement

Abstract

The challenge of demographic change combined with decreasing numbers of care personal is widely known. To ensure the service of individual care for seniors, researchers work on developing digital devices to support the caregivers in their day-to-day tasks. Especially the Covid-19 pandemic emphasized this need for flexible services. In this context, service robots are not only able to overcome this challenge, but to also support in other use cases such as therapy of children with autism spectrum disorder, or teaching. Regardless of their area of application, their usage and benefits highly depend on the user acceptance. Consequently, many studies on their appearance, behavior as well as their usefulness and acceptance have been undertaken. However, the studies on acceptance measurements are difficult to compare due to different study designs and therefore are hard to build upon. This is the case since traditional acceptance models such as the Technology Acceptance Model by Davis (1989), are not considered as sufficient for the interaction-focused technology of robots. Therefore, it is difficult and time-consuming for researchers to determine the most appropriate model in order to measure the acceptance of the developed service robots. To be able to measure new robot developments in this rapidly evolving research field, an overview on existing models as well as their application options is necessary.To support researchers and developers the aim of this research is to provide an overview on existing models for the acceptance measurements of service robots. To reach the objective, the following main research question is proposed: Which models currently exist to measure the acceptance of different service robots?Following the taxonomy on literature reviews by Cooper (1988), the focus of this literature review is on existing research methods and practices. The goal is to integrate existing literature within a matrix to identify central issues, by having a neutral perspective. The research follows the guidelines of Brocke et al. (2009) including a concept-centric approach of Webster and Watson (2002) for the literature analysis and synthesis. The foundation builds a search based on the search string robot* AND accept* AND (measur* OR method* OR model* OR evaluation*) which was used for Scopus, Science Direct, IEEE Xplore as well as Google Scholar. From the 274 identified research paper duplicates and non-English ones are excluded, which is resulting in 226 unique research paper. Those are further clustered and the 19 identified acceptance models for service robots are brought into relation to the robotic classification of Onnasch et al. (2020). The concept-matrix reveals that most models are highly specific to a certain field of application or target group. In addition, a large number of acceptance models for service robots were found, but most of them are not commonly used. This concept-centric literature review gives a structured overview that can be used by researchers and developers to quickly identify the most suitable model for their research. However, some characteristics are not covered or just covered from different models. Consequently, further research on how to overcome these gaps is required.