The work in this sub-activity focuses primarily on simulation models for the development of autonomous navigation techniques for robotic platforms operating in smart hospitals. The main challenge in this research field is to devise navigation strategies that are at the same time efficient and safe, within an environment populated by human agents.
The literature on social navigation has recently addressed this problem with a variety of solutions, combining different human motion models with a number of robot navigation strategies, either model-based or leveraging various versions of reinforcement learning approaches. A key requirement of the design process is the availability of reliable simulators for pedestrians moving in indoor environments.
The aim of this sub-activity is to develop a simulation environment integrating a generator of realistic trajectories of pedestrians moving within an hospital with a hierarchic path-planning technique for mobile robots navigating in a crowded workspace. This activity is tightly interconnected with that concerning mobile autonomous platforms, carried out in collaboration with the University of Pisa (Activity A9.3 a2).

Tuscany Health Ecosystem
SPOKE 9: Robotics and Automation for Health