Soil-dwelling pests, such as Phyllophaga larvae, pose significant challenges to agriculture as they feed on crop roots, causing substantial losses. Their hidden behavior within the soil further complicates monitoring and control efforts. Traditional methods, such as manual excavation and acoustic detection, are often invasive, labor-intensive, or limited in precision. To address these challenges, this study aimed to establish a reliable methodology to study larval trajectories and responses within the soil environment without disturbing its natural behavior. This study describes the development of an innovative system for precise tracking of these larvae, combining magnetic markers with an array of magnetoresistive sensors. Larvae were tagged with neodymium magnets and tracked using an array of 64 anisotropic magnetoresistive sensors while being attracted by food sources and repelled by electrical stimuli. The movement of larvae marked with magnetic tags and attracted by Zea mays and Solanum tuberosum roots was successfully monitored. The system was validated using a 3D printer framework as a reference, achieving high accuracy with minimal uncertainty. Adjustments were made to the z-axis to account for variations in the distance between the magnet and the sensor array. Experiments demonstrated the ability to guide larval behavior through controlled electrical stimuli, confirming the system's utility for monitoring and behavioral studies. This approach offers significant improvements over traditional methods by preserving soil integrity, enhancing precision, and enabling real-time tracking. The findings provide a valuable tool for understanding subterranean pest dynamics and support the development of sustainable pest management strategies in agriculture.