Mobile robots are widely used in logistics tasks within manufacturing automation, encompassing activities such as carrying, delivering, and receiving goods. However, as their interaction with human operators and the environment becomes more prevalent, ensuring safety in their operation becomes a paramount concern. This paper focuses on collision issues as a critical safety consideration. The function-centric hazard identification approach (F-CHIA) is employed to systematically identify hazards and their associated properties, leading to the derivation of essential safety requirements. To tackle problems of estimating collision risks and to meet safety requirements, the paper introduces a novel monitor algorithm based on the Kalman Filter (KF). In contrast to conventional risk estimation methods like risk matrices, this algorithm continuously calculates risk by leveraging trajectories and posing uncertainties predicted by the KF, utilizing real-time positions as input. The effectiveness of the algorithm is demonstrated in a simulated environment created in CoppeliaSim, showcasing its applicability in scenarios where humans and mobile robots collaborate in the same space, such as a warehouse.