Evaluation of LSRB Pathfinding Performance in an Autonomous Obstacle-Avoiding Robot

This study presents an evaluation of the performance of the Left-Straight-Right-Back (LSRB) algorithm implemented in an autonomous obstacle-avoiding robot. The LSRB algorithm operates based on a fixed priority rule in pathfinding decisions: turn left, go straight, turn right, and finally perform a 180-degree turn if no paths are available. The robot is equipped with ultrasonic sensors and a servo motor to scan obstacles on the left and right sides, and utilizes an 8×8 dot matrix display to indicate its navigation status. Testing was conducted in a custom-built maze environment featuring branches, dead ends, and narrow paths to simulate real-world navigation scenarios. Performance evaluation parameters include travel time, number of maneuvers, and path accuracy. Experimental results show that the LSRB algorithm achieved 100% path completion accuracy across all test cases, with consistent travel time and efficient obstacle avoidance. The findings demonstrate that LSRB is a reliable and lightweight navigation strategy, particularly suitable for low-cost, microcontroller-based robots used in educational or semi-structured environments. Limitations regarding power supply stability and the absence of memory-based path tracking are also identified, offering opportunities for future improvements. Keywords - Obstacle-Avoiding Robot, Robot Navigation, LSRB Algorithm, Ultrasonic Sensor