During the Covid-19 pandemic, hospitals faced challenges in coping with the virus’s rapid spread. Collaborative multi-robot systems can help make hospital services more efficient and reduce human interactions. This study describes a multi-agent reinforcement learning (MARL) approach for the deployment of mobile robots in a hospital setting, with an emphasis on adaptability to dynamic environments. The protocol includes an exploration phase, Q-learning algorithm implementation, different MARL architectures (centralized, decentralized, and independent learning), and a collaborative learning aspect using a consensus algorithm.