Solar tracking is essential for efficiency, yet current market solutions present a trade‐off: Active systems are precise but expensive and computationally demanding, while low‐cost passive systems (e.g., LDRs) often lack accuracy. To address this gap, this work presents the design and validation of a passive solar tracking sensor based on gnomon shadow analysis using computer vision. Unlike complex direct imaging systems, this approach utilizes lightweight geometric transformations and color segmentation to estimate solar azimuth and elevation in real time. Experimental validation against NOAA models yielded angular errors ranging from 0.155 to 59.13 mrad in azimuth and from 36.49 to 66.84 mrad in elevation. Robustness tests demonstrated that the system maintains consistency even with camera position variations, showing a maximum deviation of 9.4 mrad. Furthermore, a computational benchmark confirmed the algorithm′s efficiency, achieving a processing time of 3.69 s on a legacy low‐end device (Intel Atom), which is sufficient for solar dynamics. These results distinguish the proposed method as a feasible, cost‐effective alternative that achieves functional accuracy without the need for high‐performance computing resources or expensive optical encoders.
Ordaz-Castillo et al. (2026) studied this question.