UV-Based Celestial Navigation for Straight-Line Control in Outdoor Mobile Robots

Abstract—Design of control systems for outdoor vehicles typically involves sensor suites with significant weight, computational, and power requirements. One biological strategy to circumvent this complexity is observed in moths, which maintain fixed orientation relative to distant celestial light sources for straightline navigation over long distances. Since the sun and moon provide the dominant ultraviolet (UV) radiation in outdoor environments, this principle can be emulated using lightweight UV sensors to estimate and maintain a robot’s angular bearing relative to these distant reference points. This work demonstrates UV-based celestial navigation using a differential drive robot equipped with four orthogonally-mounted UV sensors. The system identifies the sensor pair receiving maximum UV intensity, computes a normalized directional reading, and maintains this reference using proportional control applied to differential wheel speeds. Experimental validation involved driving the robot along a straight concrete path for 10 s intervals, measuring horizontal deviation from the initial position with and without UV-based control. Results from 10 trials per condition showed that UV control reduced mean absolute horizontal deviation by 48 % compared to open-loop operation (20.4 inches vs. 39.1 inches, p ¡ 0.05), even under partly cloudy conditions. These findings demonstrate that intensity-based UV sensing provides measurable improvement for straight-line navigation tasks while using minimal hardware—each sensor weighs approximately 7 grams with negligible power draw. The approach offers a lightweight, bio-inspired alternative to GPS-IMU fusion or vision-based methods for applications where straight-line control suffices.