Multi-insect visual tracking and system identification techniques for inflight feedback interaction analysis

Abstract

Individual insects flying in crowded assemblies perform complex aerial maneuvers by small changes in their wing motions. To understand the individual feedback rules that permit these fast, adaptive behaviors in group flight, a high-speed tracking system is needed that is capable of simultaneously tracking both body motions and these more subtle wing motion changes for multiple insects, extending tracking beyond the previous focus on individual insects to multiple insects. In this system, we have extended our capability to track multiple insects using high speed cameras (9000 fps). To improve the biological validity of laboratory experiments, we tested this measurement system with Apis mellifera foragers habituated to transit flights through a test chamber. Processing steps consist of data association, hull reconstruction, and segmentation. An analysis based on multiple flight trajectories is presented, including the differences in flight in open and confined areas containing multiple insects and the differences due to ethanol treatment. A system identification framework applicable to extracting the interaction rules in multi-agent insect trajectories is developed.