Researchers from the University of Southern California use AI and advanced mathematical modeling to help preempt poaching attacks in protected forest ranges
For the longest time, the only way to counter the ongoing threat of illegal poaching of endangered animal species in forests and natural reserves has been experience-based human patrolling. With numbers of endangered species continuing to dwindle at alarming rates (for example, from 60,000 tigers a century ago only about 3,200 now remain,) forest officials are turning to more effective approaches to counter this threat.
Protection agencies are now working with the National Science Foundation (NSF) and the Army Research Office to utilize artificial intelligence (AI) and computer models that study cooperation and conflict to help extrapolate and predict behavior. These approaches aid in determining countermeasures and containment strategies for real-world threats such as poaching.
Milind Tambe, the professor who leads a team in the computer science department of the University of Southern California (USC) is utilising a branch of AI called game theory, which is based on mathematical models that study past patterns to help predict onward behaviour of adversaries within a given system--in this case the complex functioning of wildlife reserves.
"This research is a step in demonstrating that AI can have a really significant positive impact on society and allow us to assist humanity in solving some of the major challenges we face," Tambe said.
In 2013, his team created the first AI-based program to be applied in this field, called PAWS (Protection Assistant for Wildlife Security.) It has been tested in Uganda and Malaysia in 2014, where pilot testing has led to improvements in its efficacy.
The system uses big data from numerous real-world variables including terrain data, area topography and natural transit paths of animals, in order to generate optimal routes for patrols to traverse quicker, utilising fewer resources.
The more data the system accumulates, the more adept it becomes at refining its findings and suggesting more effective ways of monitoring migratory patterns as well as predicting and countering poaching activity. The system is also able to intelligently ‘randomize’ itself to make it more difficult for poachers to predict.
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
 "Researchers patrolling a Malaysian rainforest as part of the PAWS program")
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}
)
){kind=small}
){kind=small}
)
)
)
)
)
){kind=small}
){kind=small}
){kind=small}
){kind=small}
){kind=small}