UC Santa Cruz

Animal tracking tags are a useful tool in the study of wild animals and ecological systems, but the process of tagging an animal is costly and often traumatic for the animal. For many types of tags, the lifetime of the tag is determined by battery life. The Energy Scavenging Collar for Animal Physiology and Ecology (ESCAPE) project investigates several methods for reducing overall power consumption in wildlife tracking tags. This work develops and tests several algorithms for reducing GPS uptime in these tags, using Extended Kalman Filters to guide uncertainty-suppression GPS scheduling. These algorithms use less costly sensors such as accelerometers, magnetometers, and gyrometers to augment GPS in estimating locations during tracking. We implement and test these algorithms in tests on human subjects. This work also includes analysis of several proposed techniques for scavenging energy in the field, such as harvesting excess kinetic and thermal energy from the animal. Our analyses provide upper bounds on expected power returns from these strategies, providing useful metrics for determining whether such strategies are viable regions of the tag design space. The findings of the ESCAPE project contribute to future tag research and development by highlighting strategies with potential for improving tag lifetimes, and discouraging other strategies.