20.500.12556/RUL-182360 Constraint-driven neural network architecture adaptation for image localisation on a miniature mobile robot Deploying deep learning models on low-power devices is an increasingly important research challenge. Microcontrollers, in particular, have strict limitations due to the limited processing power and memory. Specifically, object localisation on such devices remains in the early stages of research, showing the need for architectural designs tailored to resource-constrained environments. In this work, we implement and evaluate different architectural configurations of deep learning models on a miniature mobile robot operating in a granular, sand-like terrain, where both robots and rocks must be localised as a prerequisite for robust navigation. The localisation models are based on the MobileNetV2 architecture utilising the FOMO (Faster Objects, More Objects) algorithm with dynamic convolutions. The proposed system consists of an HM01B0 greyscale camera integrated with a low-power RP2040 microcontroller. The object detection models are trained on a custom dataset consisting of 10,000 greyscale images (160×160 px) featuring robots and rocks. Our findings indicate that the best trade-offs between accuracy and memory are achieved by truncating the middle layers of the base model, balancing output resolution with retained pre-trained features and classifier complexity. Additionally, the results indicate that increasing the input image resolution produces a greater improvement in accuracy compared to expanding the model’s classifier head, relative to the increase in inference time. This study presents the first large-scale variation of FOMO truncations for the purpose of object localisation on low-power microcontrollers and offers practical guidance for selecting neural network architectures. convolutional neural network low-power microcontroller miniature mobile robot MobileNet object localisation tiny machine learning TinyML konvolucijske nevronske mreže nizka moč mikrokrmilniki miniaturni roboti MobileNet lokalizacija objektov TinyML true false true Angleški jezik Slovenski jezik Članek v reviji 2026-05-08 09:14:04 2026-05-08 09:14:12 2026-05-09 04:44:15 0000-00-00 00:00:00 2026 0 0 Str. 13497-13506 no. 9 Vol. 26 1 May 2026 0000-00-00 Zaloznikova Objavljeno apc 0000-00-00 0000-00-00 0000-00-00 681.527.7:004.4 1530-437X 10.1109/JSEN.2026.3674225 275091971 Pleterski_..._Constraint-Driven...pdf Pleterski_..._Constraint-Driven...pdf 1 41F21AD7E0DAFD67098C22F58150A868 0804e5bea41511756b7239ce058ea89ec2a1403cf8ab77b928fc9ea7837e160c 76a22c75-4aad-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=233529 https://ieeexplore.ieee.org/document/11449403 1 3e95cefe-4aad-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=233528 Fakulteta za strojništvo 0 0 0