Computer scientists from the University of Southern California (USC) developed an artificial intelligence tool that helps prevent poaching by predicting where it is likely to happen and suggesting patrolling routes.
Elephants, lions and rhinoceros have all been in the news lately, but for a sad reason: poaching. They are being hunted for several reasons: ivory trade, game hunting/trophies, food, and more. These species we have all grown to love might disappear if poaching continues unchecked.
Faced with this grim reality, Milind Tambe, a computer science professor from USC, decided to create a team of PhD students to find a solution. They collaborated with other researchers and wildlife organisations. This exercise, which applied principles of game theory, resulted in an artificial intelligence tool called PAWS: Protection Assistant for Wildlife Security.
“We can do pattern recognition, essentially building a model and predicting poachers’ activity based on their past actions. We can build up predictions of where poachers may strike. Then that allows us to generate patrol strategies that would be effective against those types of poachers. ”
– Milind Tambe
PAWS generates route suggestions for patrollers. Its algorithm takes into account not only animal densities and previous poaching indicents, but also topography, elevation and hydrography. Based on all this information, PAWS then generates maps with easy routes for the patrollers that maximise their changes of catching poachers. The routes are randomised to increase the change of catching poachers unexpectedly.
The more patrollers use PAWS, the better it gets. As new poaching incidents are observed and recorded, the tool’s predictions become more accurate.
PAWS has already been tested in Malaysia and Uganda, and has received positive feedback so far.