<p>In real-world scenarios, mobile service robots maneuver safely and effectively in human environments, which poses a significant challenge. The Social Force Model (SFM) is an effective socially aware navigation algorithm for robots in crowded environments. However, it has limitations, particularly in handling dynamic obstacles and moving humans. The model primarily considers the distance between the robot and obstacles or humans, neglecting their velocities. This oversight can lead to collisions with moving humans and dynamic obstacles. Additionally, the velocities produced by the original SFM can change suddenly, resulting in instability that compromises safety and predictability. This paper proposes a Modified Social Force Model (MSFM) that considers the velocity and threshold distance of obstacles and humans to enhance the robot's navigation performance in complex environments. Additionally, a PID controller was combined with the MSFM to improve the stability of the robot's velocity. The effectiveness and feasibility of the proposed MSFM were investigated through simulation using MATLAB and ROS framework, under different environmental conditions in three simulated maps. The results indicated an average reduction of 8.76% in path length and 20.55% in root-mean-square error of the robot’s velocity for three cases when comparing the MSFM with PID controller to other human-aware navigation algorithms such as SFM, hybrid reciprocal velocity obstacles. This illustrated that the proposed method can fully drive a mobile robot safely and efficiently in complex situations.</p>