CHENNAI: An algorithm that can help combat wildlife poaching by providing efficient patrolling using available limited resources has been developed by researchers from Indian Institute of Technology Madras and Harvard University.
Called CombSGPO or Combined Security Game Policy Optimisation, it provides strategy for resource allocation and patrolling for protecting wildlife using data on the animal population in the conserved area. It assumes poachers are aware of the patrolling being done at various sites.
“The work was motivated by the need to perform strategic resource allocation and patrolling in green security domains to prevent illegal activities such as wildlife poaching, illegal logging and illegal fishing,” said Prof Balaraman Ravindran, head of IIT-M’s Robert Bosch Centre for Data Science and Artificial Intelligence. “The resources we consider are human patrollers (forest rangers) and surveillance drones, which have object detectors mounted on them for animals and poachers and can perform strategic signalling and communicate with each other as well as the human patrollers,” he added.
The World Wide Fund for Nature (WWF) says wildlife trade poses the second biggest direct threat to the survival of species after habitat destruction. A study by Traffic, a wildlife trade monitoring network of the WWF, showed that more than 1 lakh tortoises/freshwater turtles have been illegally traded across India since 2009.
The study by Prof Ravindran’s team and Prof Milind Tambe’s research group ‘Teamcore’ at Harvard University realised that allocating resources of rangers and drones with coordinated patrolling with real-time communication can be a good strategy to protect wildlife in conserved areas. But none of the earlier models included these components.
Prof Ravindran said the model was based on game theory that works in two steps. First, the algorithm handles resource allocation and in the second stage it strategizes patrolling after the allocation has been done. Game theory tries to predict how a player will choose different strategies when the outcome depends on how everyone else in the group behaves.
Researchers said past models based on game theory predicted the areas where poaching may take place based on earlier poaching incidents and interactions between poachers and defenders. The new algorithm, when compared with similar existing tools, was found to provide better strategies and is more scalable than earlier ones, professor Ravindran said.