Modeling and simulation of obstacle detection for a mobile robot

Autonomous navigation in cluttered static environments remains a practical challenge, particularly when integrating discrete obstacle avoidance logic with continuous motion control. This paper presents a simulation study evaluating a hybrid navigation architecture for a differential-drive mobile robot in a 2D environment with static obstacles. The control strategy relies on a state machine: a discrete Proportional-Integral-Derivative (PID) controller handles normal waypoint tracking, while a 16-ray virtual sensor array triggers a reactive evasion mode based on inverse-square repulsive directional weighting when proximal threats are detected. The system’s kinematic efficiency and safety margins were quantitatively evaluated by analyzing its spatial trajectory, velocity profiles, and clearance margins. While the robot successfully navigated the 35.13-m course in 202.6 s and avoided physical collisions (maintaining a 1.00-m absolute minimum clearance, despite breaching the ideal 2.0-m safety margin), the purely reactive logic proved highly inefficient. The average speed dropped to just 0.17 m/s, and the abrupt switching between operational modes caused severe control signal chattering and prolonged stagnation in local minima. These results show that relying solely on local reactive rules heavily penalizes performance in confined spaces. Consequently, from a theoretical and geometric standpoint, the limitations observed in this study suggest that improving navigation efficiency in future implementations would benefit from coupling the local PID loop with a global predictive path planner, such as A* or Rapidly-exploring Random Trees (RRT), to anticipate geometric traps.