In a recent study published in the journal Scientific Reports, researchers introduced a new benchmarking method to evaluate the autonomous navigation skills of mobile robots in hospital settings.
The tested robotic platforms: HOSBOT (a) and TIAGo (b). Image Credit: https://www.nature.com/articles/s41598-024-69040-z
The goal was to establish a standard protocol and performance metrics for assessing these robots in complex clinical environments. This method aims to fill the gap in the current literature caused by the lack of standardized evaluation methods for autonomous mobile robots (AMRs) in such settings.
Background
The use of robotics in healthcare has seen significant progress in recent years, providing various solutions to enhance the effectiveness, quality, and accuracy of medical procedures and patient care. These robotic devices are developed to automate manual repetitive tasks, boost efficiency, reduce costs, support clinical staff, and improve patient experiences. They perform a range of functions, including delivering drugs and supplies, cleaning and disinfecting, and assisting in rehabilitation.
However, the widespread use of these robots highlights concerns about their reliability and user safety, particularly in unstructured and sensitive environments. Hospitals are complex settings with crowded hallways, waiting areas, and patient rooms filled with medical staff, patients, visitors, and equipment. Developing robots that can navigate autonomously in such environments is challenging. These robots must detect and avoid collisions, predict human movements, and plan paths to minimize risks to both them and people.
About the Research
In this paper, the authors proposed a structured benchmarking method to quantitatively assess the autonomous navigation capabilities of mobile robots in hospital environments. This method includes a protocol with standardized test batches and performance metrics to comprehensively evaluate the robots' ability to navigate around dynamic and static obstacles and interact with humans.
The benchmarking protocol consists of four test batches with increasing path complexity, from simple linear movement to complex scenarios with various obstacles. The performance metrics include completion time (CT), path length (PL), deviation error (DE), orientation error (OE), success rate (SR), and minimum distance from the obstacle (MDO). These metrics assess the robots' ability to reach destinations, maintain safe distances from obstacles, and adjust their paths in changing conditions.
Additionally, the researchers tested their method on two AMRs: HOSBOT, a robot for hospital logistics, and TIAGo, a robot with an anthropomorphic arm. The experiments were conducted in a simulated environment that replicated real hospital conditions, including patient beds, corridors, waiting areas, and furnishings.
Research Findings
The outcomes demonstrated that the proposed benchmarking method effectively measured the autonomous navigation performance of the tested robots. HOSBOT and TIAGo successfully navigated the hospital environment, even with dynamic and static obstacles, while maintaining a high SR and avoiding collisions. However, the performance indicators, such as CT, PL, and position/OEs, tended to deteriorate as the speed and complexity of the environment increased.
The statistical analysis revealed that as the environment's complexity and the robots' speed increased, their accuracy in reaching target positions and orientations decreased. Despite this, the robots managed to adjust their paths in real time to avoid obstacles and reach their destinations. The study also emphasized the need to consider the robot's size relative to the environment, as this affects maneuverability and navigation performance.
Applications
The proposed benchmarking method is designed to assess whether AMRs are suitable for hospital environments and to ensure they meet required safety and performance standards. By employing standardized tests and performance metrics, this method can highlight the strengths and weaknesses of various robots, facilitating the selection of the most appropriate ones for specific tasks such as logistics, disinfection, or rehabilitation assistance.
Furthermore, the benchmarking data can inform the creation of guidelines and regulations for medical robots with autonomous navigation capabilities. This information will be invaluable in guiding the design and deployment of these robots, ensuring that they prioritize the safety and well-being of patients, staff, and visitors.
Conclusion
In summary, the novel benchmarking method proved effective for evaluating the autonomous navigation capabilities of mobile robots in hospital environments. It could help assess the reliability and safety of robotic platforms for various clinical applications. Additionally, it could support the development of guidelines and best practices for integrating these robots into healthcare settings, ultimately enhancing patient care quality and improving clinical efficiency.
Future work should focus on collecting data from robots using different navigation algorithms to evaluate their impact on performance and technical requirements. This includes refining the benchmarking protocol and expanding performance indicators for a better understanding of robots' capabilities in real-world hospital settings. The authors also suggested exploring human-robot interactions, including how human behavior influences robot navigation, designing robots for safe and effective human interaction, and developing algorithms that adapt to dynamic and unpredictable environments.
Journal Reference
Rondoni, C., Scotto di Luzio, F., Tamantini, C. et al. Navigation benchmarking for autonomous mobile robots in hospital environment. Sci Rep 14, 18334 (2024). DOI: 10.1038/s41598-024-69040-z, https://www.nature.com/articles/s41598-024-69040-z
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.