Robotic Monitoring of Bird Presence in Pantanos de Villa Wildlife Refuge
Luciana Rojas
31/08/2026
Wetlands are among the most productive and ecologically important ecosystems on Earth, yet they are increasingly threatened by urban expansion, pollution, and habitat degradation, particularly at Pantanos de Villa Wildlife Refuge (PVWR), a coastal Ramsar wetland surrounded by the rapidly growing city of Lima, Peru. Birds are widely recognized as effective bioindicators of wetland conditions because their presence, abundance, and distribution respond rapidly to changes in habitat quality, water levels, and food availability. However, conventional bird monitoring relies on manual observation, which is costly, temporally limited, and can disturb wildlife.
This study investigates whether a low-cost autonomous robotic platform can address these limitations by recording bird presence with minimal disruption. A robotic monitoring system was constructed from low-cost recycled materials and 3D-printed connectors, using an Arduino Mega, an HC-SR04 ultrasonic sensor, and an ESP32-CAM to autonomously detect movement and capture images without human supervision. The system was evaluated through controlled testing, a field deployment at PVWR, and a benchtop characterization of the ultrasonic sensor.
During controlled testing, the system consistently logged detection events. The ultrasonic sensor achieved 100% detection reliability within its main lobe, decreasing to 70–90% at the fringe and 10–40% within the side lobes. The camera successfully captured an image for every valid sensor detection. Although no confirmed bird detections were recorded during the single field trial, despite birds being present nearby, the results demonstrate the technical feasibility of a low-cost, non-invasive autonomous bird-detection platform. However, this work represents a proof of concept rather than a validated environmental-monitoring tool, and further field validation alongside independent environmental measurements is needed to establish bird presence as an indicator of wetland condition.