This research is focused on the performance of a Deep Reinforcement Learning method on an agent (mobile robot) in a simulated virtual environment (Operating Room) for medical applications. The purpose of this research is to compare suitable decisive actions taken by the agent to achieve its goal target. Executing this goal requires the implementation of a reward-penalty system for observation and analysis. The agent’s accumulated reward is based on the best-navigated decision to avoid collisions; solely generating an intelligent agent system. We reviewed previous works on the impact of Deep Reinforcement Learning algorithms on an agent in areas of navigation and exploration. Adopting a Deep Reinforcement Learning method and a physical simulator, we trained and tested the agent using existing environments and our modeled operating room, respectively. Measuring the positive reward output of the experiment with different parameters of the algorithm such as the learning rate, maximum Q-value and the average time to attain its goal position, we presented our work with plots of the experiment and compared it with a widely known traditional method. Our experimental results indicated that the agent achieved a high positive reward of 3800 in our operating room environment with a learning rate of 0.5. Our research aimed at training an agent to make intelligent decisions in achieving its goal destination without prior experience and input data. Reinforcement Learning provides a structure for robotics to function effectively; utilizing and engaging a robot to navigate and explore in any given environment.