Older adults living alone are a highly vulnerable population for whom delayed response to medical emergencies increases the risk of serious health consequences and contributes significantly to healthcare costs. The recent availability of low-cost robots (<€5,000) capable of autonomous navigation in domestic environments, acquiring audio/video/3D data streams, and communicating with Large Multimodal Models enables the exploration and experimentation of novel and promising scenarios for monitoring and assistance of such elderly individuals and, more in general, for persons with limited autonomy living independently at home. Within this context, as part of the PNRR project “Age-It: Ageing Well in an Ageing Society,” the Bari Research Unit is investigating the deployment of the Unitree Go2 robot to enhance the physical safety of frail patients in domestic environments and residential care facilities. However, the procurement logistics for these emerging robots of Chinese or US origin present significant challenges, including extended waiting periods of several months, complex customs and administrative procedures, and the substantial inability to conduct hands-on verification of the robots’ actual capabilities prior to purchase. To address these challenges and validate the Go2's applicability in our research scenarios, we employed the Nvidia Isaac Sim simulator within the open-source ROS2 (Robot Operating System, v.2) environment to assess whether the Go2’s specifications—specifically its LiDAR system, RGB-D camera, audio system, inertial sensors, and autonomous navigation capabilities—were suitable for our intended use cases. Validation through simulation not only provided positive confirmation of the robot's characteristics but also enabled us to accelerate the prototyping timeline for Go2 control software, data collection, LMM interaction, data analysis, and emergency alert activation systems. This paper provides a detailed description of the use case scenario, design decisions, and achieved results.