Eduventure: instruction manual and educational kit for ai-based autonomous line finder robot and ball finder robot leveraging webots/ Farah Louise A. Berin, Denisse Roxanne H. Bolinas, Miguel S. De Leola, Ronald James R. Faigao, Randell James F. Mendoza, and Ronald France J. Percela.--

Bachelor's thesis

College Of Engineering.--
Bachelor of science in electronics engineering: Technological University of the Philippines,
2025.

Includes bibliographic references and index.

This study addresses the gap in accessible robotics education within resource-limited
academic settings, specifically at the Technological University of the Philippines Manila (TUPM),

where a lack of modern tools hinders comprehensive learning. We present Eduventure, an AI-
integrated autonomous mobile robot educational kit featuring a Line Finder Robot and a Ball Finder

Robot, both powered by Raspberry Pi and simulated in Webots. The Line Finder Robot demonstrated
high consistency and reliable performance across six line patterns derived from Dewantoro et al.
(2021), achieving low mean traversal times (e.g., Pattern 1: 7.87 s, SD=0.23). The Ball Finder Robot
showed consistent and scalable object detection, with mean detection times of 1.75 s at 2 feet, 3.16
s at 4 feet, and 4.89 s at 6 feet. To assess educational value, a survey of TUPM students revealed
high satisfaction (average scores 3.53-4.70) regarding clarity, usability, and educational impact,
praising the kit's comprehensibility and hands-on engagement. Qualitative feedback highlighted
needs for improved guidance and troubleshooting. Overall, our results demonstrate that low-cost,
AI-powered robotics kits can provide effective, engaging, and accessible learning experiences in
robotics and AI, particularly for institutions with limited access to advanced laboratory facilities.
Future work will focus on enhancing instructional materials, simulation realism, and integrating
advanced object detection models like YOLO.