Elemantra: An End-to-End Automated Framework Empowered with AI and IoT for Tackling Human-Elephant Conflict in Elephant-Range Countries

Nuwan Sriyantha Bandara, Dilshan Pramudith Bandara
IEEE Sensors Letters [In review]
Paper Code EleThermal Dataset

Abstract

The cohabitation of elephants and humans has evolved into a human-elephant conflict (HEC) due to the increasing loss of historical elephant habitations. Since HEC is a substantial threat to both species, advanced sensing methods are utilized to develop HEC prevention frameworks which still lack unification. Here, we propose an end-to-end automated framework for HEC prevention consisting of three main modules: a distributed deep learning-assisted elephant detection module using infrared and seismic sensing, an on-site repelling system with time-varying acoustic and light deterrents and a mesh network for device communication. The framework is equipped with novel decision-making pipelines and algorithms such that it has the potential to operate with no human intervention. The preliminary results from each module confirm that the proposed framework is effective in performing their individual tasks towards collaboratively achieving the prevention of HEC.

BibTeX

@misc{bandara2023elemantra,
      title={Elemantra: An End-to-End Automated Framework Empowered with AI and IoT for Tackling Human-Elephant Conflict in Elephant-Range Countries}, 
      author={Nuwan Sriyantha Bandara and Dilshan Pramudith Bandara},
      year={2023},
      eprint={2310.15012},
      archivePrefix={arXiv},
      primaryClass={eess.SY}
    }