AI-enabled Wild Boar Intrusion Detection and Deterrent System
Keywords:
Edge computing, IoT, Precision agriculture, PIR sensor, Thermal detection, Ultrasonic deterrent, Wild boar intrusion, YOLOv8Abstract
Agriculture in forest-adjacent regions has long struggled with wildlife intrusions, and wild boar attacks are among the most destructive. Farmers often suffer serious crop losses, yet the tools available to them—electric fences, chemical repellents, or manual night patrols—demand continuous effort and still fall short in terms of reliability. This study presents an AI-driven intrusion detection and deterrent system specifically designed to tackle wild boar incursions in an automated, humane, and energy-efficient manner. The system employs a dual-sensor approach using passive infrared (PIR) and thermal sensors as the first line of motion detection. When a potential intrusion is triggered, a Raspberry Pi 4 running a custom-trained YOLOv8 model analyzes the captured frames to confirm the presence of a wild boar. Upon verified detection, an ultrasonic deterrent module emitting frequencies between 21–40 kHz activates to drive the animal away without physical harm. At the same time, real-time push notifications are sent to the farmer through “Boarex”—a custom Flutter mobile application—even when the app runs in the background. Every event is timestamped and logged, helping farmers track intrusion patterns and plan better protective strategies over time. Experimental testing demonstrated a detection precision of 94.2% and a recall of 91.6% for wild boars. The system processes video at 12–15 FPS with an inference delay of approximately 280 ms. False positives dropped by over 90% compared to conventional PIR-only setups, and power consumption in event-driven mode was around 2.1 W—roughly 67% less than continuous monitoring. The solar-powered design makes this solution viable even in remote, off-grid farmlands.
References
I. N. Simasiku, B. J. Temu, and G. Z. Nyamoga, “Assessment of conflicts under human-wildlife interactions: An application of the conservation conflict transformation model in communities adjacent to Nyerere National Park, Tanzania,” Science of the Total Environment, vol. 951, Art. No. 175890, Nov. 2024.
A. Joshi, N. Pandey, M. Diwakar, P. Singh, A. Shankar, and F. Alqahtani, “WBD-YOLO-AM: Yolov8 with attention module and IoT-based wild boar detection and deterrence system for safeguarding small millets,” Turkish Journal of Agriculture and Forestry, vol. 49, no. 4, pp. 769–786, Aug. 2025.
H. J. König, C. Kiffner, S. Kramer‐Schadt, C. Fürst, O. Keuling, and A. T. Ford, “Human-wildlife coexistence in a changing world,” Conservation Biology, vol. 34, no. 4, pp. 786–794, May 2020.
C. Conejero et al., “Between conflict and reciprocal habituation: Human-wild boar coexistence in urban areas,” Science of the Total Environment, vol. 936, Art. No. 173258, Aug. 2024.
D. Singh, A. R. Kadam, and S. B. Dethe, “Integrating ultrasonic sound technology for wild boar deterrence in agriculture: A comprehensive research analysis,” Journal of Theoretical Physics & Mathematics Research, vol. 2, no. 2, pp. 1–6, Apr. 2024.
G. H. Shoddo, “Navigating the timber nexus: Balancing livelihoods and environmental concerns in the Sheka Zone, Ethiopia,” Sustainable Futures, vol. 8, Art. No. 100252, 2024.
M. Barua, S. A. Bhagwat, and S. Jadhav, “The hidden dimensions of human–wildlife conflict,” Biological Conservation, vol. 157, pp. 309–316, 2013.
N. Abed, R. Murugan, and S. Manaili, “Optimizing synergistic combinations of adaptive IoT-based animal repellent systems for sustainable agriculture in Rajasthan, India,” Agricultural Science Digest, vol. 44, no. 3, pp. 580–587, Aug. 2024.
S. Goel and R. Sharma, “Assessment of crop damage by wild animals and renewable energy interventions—A case study from coastal Odisha, India,” in 2021 1st Odisha International Conference on Electrical Power Engineering, Communication and Computing Technology (ODICON), Bhubaneswar, India, 2021, pp. 1–5.
M. Martinský, J. Papán and S. Tatarka, “Deployment and design of AI-powered edge systems for wildlife monitoring: A survey,” 2025 International Conference on Emerging eLearning Technologies and Applications (ICETA), Stary Smokovec, Slovakia, 2025, pp. 566–571.
S. Liu, “Towards a sustainable agriculture: Achievements and challenges of sustainable development goal indicator 2.4.1,” Global Food Security, vol. 37, Art. No. 100694, Jun. 2023.
M. Hajder, J. Kolbusz, and M. Liput, “An AI-based integrated multi-sensor system with edge computing for the adaptive management of human-wildlife conflict,” Sensors, vol. 25, no. 20, p. 6415, Oct. 2025.
G. Ramesh et al., “An automated vision-based method to detect elephants for mitigation of human–elephant conflicts,” in 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Udupi, India, 2017, pp. 2284–2288.
A. Balaji, B. Sathyasri, V. V. Reddy S, D. Indumathy, R. Krishnan, and S. Vanaja, “Intruder alert system in smart home based on IoT technique,” in 2022 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS), Chennai, India, 2022, pp. 1–4.
M. McKerchar et al., “The potential for wildflower interventions to enhance natural enemies and pollinators in commercial apple orchards,” Agriculture, Ecosystems & Environment, vol. 301, Art. No. 107034, 2020.
