Mango production is a cornerstone of Burkina Faso’s agricultural economy, significantly contributing to national fruit output and economic stability. However, traditional mango harvesting methods are labor-intensive and inefficient, limiting scalability. This study explores an innovative solution combining artificial intelligence (AI) and unmanned aerial vehicles (UAVs) for automated mango harvesting. By leveraging the YOLO (You Only Look Once) algorithm for object recognition, this research compares the performance of YOLOv8s, YOLOv9s, and YOLOv10s in detecting and localizing ripe and unripe mangoes within orchard environments. The results demonstrate that YOLOv10s has the highest precision, recall, and F1-score for MangueMure (MM) means ripe mangoes at 95.6%, ensuring exceptional detection accuracy, and good balance for MangueNonMure (MNM) means unripe mangoes with all metrics at 85.4%, showing reliable detection, along with the highest overall accuracy for MM at 95.6%, indicating the most precise detection. However, it has slightly lower accuracy for MNM compared to YOLOv9s at 85.4% but remains significantly high. Integrating this AI model with UAV technology has the potential to revolutionize agricultural practices in Burkina Faso, promoting sustainability and improving food security.