Insect Group/Swarm Behaviors and their Relation to Individual Feedback Models

Abstract

Apply new insect kinematics analysis techniques to extract the strategies insectsuse in aerial maneuvering in dense, high traffic environments, including swarm behaviors. By applying new tools from control theory, dynamics modeling and system identification, and leveraging significant recent improvements in aerial multi-insect tracking capabilities, we are able to simultaneously quantify the instantaneous feedback control targets and time histories of individual organisms~ neural function during group and swarm behaviors. In order to develop quantitative models of insect behaviors, we need mechanisms that allow us to induce and identify the immediate target behavior, and then monitor the feedback regulation strategies the insect uses to regulate to that target. In this work, we rely on current understanding of insect visual response fields to apply visual stimulus and induce insect behavior targets. We apply a new technique called optimal identification (OID) that incorporates open loop dynamic modeling, closed system identification, controller estimation, and optimality conditions to recover which state variables an insect controller works to minimize (in a 2-norm sense). We measure the behavior of the insect, both in isolation and receiving group stimulus, to understand how the individual behaviors (as quantified by their optimality targets) are modified by group presence, and to identify the interactions that support a critical transition to swarm behaviors. The results of this project will provide a mathematically-rigorous set of quantitative models representing biologically-derived interactions in untethered insect group and swarm behaviors that may be translated to the computationally-limited autonomy challenges of engineering fields of small aerial robots into fast, agile, and adaptive swarms.

Document Details

Document Type
DoD Grant Award
Publication Date
Apr 24, 2019
Source ID
N000141912216

Entities

People

  • Imraan Faruque

Organizations

  • Office of Naval Research
  • Oklahoma State University–Stillwater
  • United States Navy

Tags

Fields of Study

  • Biology

Readers

  • Aerial Unmanned Vehicle Swarm Micro Periodontal Dentistry.
  • Robotics and Automation.
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Machine Learning Algorithms
  • Autonomy
  • Autonomy - Autonomous System Control