Data-Driven Multimodal Patrol Planning for Anti-poaching

• DOI: 10.1609/aaai.v35i17.17792
• Publication Date: 2022-09-08T20:21:18Z
• Authors: Weizhe Chen; Weinan Zhang; Duo Liu; Weiping Li; Xiaojun Shi; Fei Fang
• Keywords: Poaching

Wildlife poaching is threatening key species that play important roles in the ecosystem. With historical ranger patrol records, it possible to provide data-driven predictions of threats and plan patrols combat poaching. However, patrollers often a multimodal way, which combines driving walking. It tedious task for domain experts manually such as result, planned routes are far from optimal. In this paper, we propose approach planning. We first use machine learning models predict then novel mixed-integer linear programming-based algorithm route. field test focusing on prediction result at Jilin Huangnihe National Nature Reserve (HNHR) December 2019, rangers found 42 snares, significantly higher than record. Our offline experiments show resulting can improve efficiency thus they serve basis future deployment field.