2023/12/07 by Atefeh Fereydooni, Armin Azarakhsh, Fereydooni, Atefeh +7
Computer Science · Psychology · #Context-Aware Activity Recognition Systems #FOS: Computer and information sciences #FOS: Electrical engineering #Human-Automation Interaction and Safety #Robotics (cs.RO) #Social Robot Interaction and HRI #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2312.04056
openalex publication_date 2023/12/07 · openalex created_date 2023/12/09 · openalex updated_date 2026/07/28
Safety concerns have risen as robots become more integrated into our daily lives and continue to interact closely with humans. One of the most crucial safety priorities is preventing collisions between robots and people walking nearby. Despite developing various algorithms to address this issue, evaluating their effectiveness on a cost-effective test bench remains a significant challenge. In this work, we propose a solution by introducing a simple yet functional platform that enables researchers and developers to assess how humans interact with mobile robots. This platform is designed to provide a quick yet accurate evaluation of the performance of safe interaction algorithms and make informed decisions for future development. The platform's features and structure are detailed, along with the initial testing results using two preliminary algorithms. The results obtained from the evaluation were consistent with theoretical calculations, demonstrating its effectiveness in assessing human-robot interaction. Our solution provides a preliminary yet reliable approach to ensure the safety of both robots and humans in their daily interactions.