Social Robots and Edge Computing: Integrating Cloud Robotics in Social Interaction
摘要
In this study, the authors explore the integration of cloud robotics for social interactions using the SoftBank Robotics Pepper robot in relation with an academic environment. The main objective is to enhance social robots ability to interact with humans by providing guidance and information using the Pepper robot as a case study. The initial steps assess Pepper’s language processing capabilities, identifying limitations in its vocabulary. The novelty of this research lies in employing Meta LLAMA 2, a large language model, trained on educational content, to enhance audio-to-text coversion. This approach aims to boost Pepper’s comprehension and response to human inquiries, marking a step forward in the practical application of cloud-edge robotics in social contexts.