SBIR Phase II: Digital Biosecurity for Invasive Insect Pests

Farmsense Inc., Riverside CA

Investigators

Abstract

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project will be in protecting natural and commercial resources as a result of actionable information on invasive insect pest. The information will be used by local, state and federal agencies to monitor harmful pests in real time and to plan optimal intervention strategies. Invasive species affect vulnerable communities in poor rural areas, who depend on natural resources, healthy ecosystems, and tourism for their livelihoods. Invasive insect pests can drive food insecurity and undermine ongoing investments in development. There is a growing consensus that early detection and rapid response is the best solution, but early detection is challenging for want of robust, automatic surveillance systems. This Small Business Innovation Research (SBIR) Phase II project will investigate problems at the intersection of entomology and engineering to improve the accuracy and reliability of autonomous insect traps. New classification models will be created to allow application to a more diverse set of insects. Novel “cold start” algorithms will be created to allow the classification of new species without first obtaining detailed training data. The project will create a scheduling system for trap servicing that dynamically considers the probabilistic insect classification/counts from thousands of traps and the location of a fleet of trap technicians (in their service trucks), in order to compute the optimal trap-to-technician assignment and order of visitation. These algorithms generalize fleet dispatch algorithms to consider the probabilistic nature of the classification and different costs of damage created by different insects. The proposed innovation of this project will lie in de-skilling the invasive species monitoring task, decreasing its cost, and reducing the lag between an insect species’ arrival and its (human) detection. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.