Edge-AI Platform for Realtime Wildlife Repelling
| Title | Edge-AI Platform for Realtime Wildlife Repelling |
| Publication Type | Conference Paper |
| Year of Publication | 2022 |
| Authors | Tamburello, Marialaura, Caruso, Giuseppe, Giordano, Stefano, Adami, Davide, Ojo, Mike |
| Conference Name | 2022 IEEE 21st Mediterranean Electrotechnical Conference (MELECON) |
| Date Published | June |
| Keywords | 6L0WPAN, 6LoWPAN, Acoustics, cloud computing, composability, controller area network, Ecosystems, edge computing, IoT, Machine Learning., pubcrawl, Real-time Systems, resilience, Resiliency, Time-series Database, ultrasonic imaging, Wildlife, Wireless sensor networks |
| Abstract | In this paper, we present the architecture of a Smart Industry inspired platform designed for Agriculture 4.0 applications and, specifically, to optimize an ecosystem of SW and HW components for animal repelling. The platform implementation aims to obtain reliability and energy efficiency in a system aimed to detect, recognize, identify, and repel wildlife by generating specific ultrasound signals. The wireless sensor network is composed of OpenMote hardware devices coordinated on a mesh network based on the 6LoWPAN protocol, and connected to an FPGA-based board. The system, activated when an animal is detected, elaborates the data received from a video camera connected to FPGA-based hardware devices and then activates different ultrasonic jammers belonging to the OpenMotes network devices. This way, in real-time wildlife will be progressively moved away from the field to be preserved by the activation of specific ultrasonic generators. To monitor the daily behavior of the wildlife, the ecosystem is expanded using a time series database running on a Cloud platform. |
| DOI | 10.1109/MELECON53508.2022.9843117 |
| Citation Key | 9843117 |