Mobile robot navigation using theta star path planning artificial potential field avoidance pure pursuit tracking and LIDAR based occupancy grid mapping

Reliable autonomous navigation of ground mobile robots in cluttered environments requires coordinated global path planning, local reactive obstacle avoidance, path-tracking control, perception, and mapping. This paper presents an integrated MATLAB-based simulation framework for a four-wheel mobile robot combining Theta* for global path planning, an Artificial Potential Field (APF) repulsion term for local obstacle avoidance, a Pure Pursuit controller, a simulated 2D LIDAR sensor, and ray-casting-based occupancy grid mapping. The framework further includes stall detection and automatic Theta* replanning to recover from local-minimum or dead-end situations. The system was evaluated in an 11.0 m 8.5 m workspace containing 22 static obstacles, across 13 trials with randomly chosen start/goal pairs; nine reached the goal, with a mean cross-track deviation of 6.9 cm and mean minimum clearance of 0.29 m, and no stall-triggered replan occurred. Follow-up experiments show that removing APF roughly halves completion and increases collisions; that a standard A* planner matches Theta*’s path length within 1% and outperformed it on one case; and that completion falls sharply in a denser layout (1 of 4 trials) versus a sparser one (4 of 4). A forced-trigger test separately confirms the stall-recovery code functions correctly when invoked. These results indicate that combining an any-angle global planner with a lightweight reactive layer and a stall-triggered replanning mechanism is a promising, computationally inexpensive approach to navigation in cluttered static environments, though its advantages over simpler baselines and its robustness to denser clutter appear environment-dependent rather than general, warranting further validation.