A human-machine symbiotic system for the extraction of high-level behaviors from a macroscopic view of swarms
Abstract
As future Air Force missions focus on robustness, cost-effectiveness and scalability, the swarming paradigm has gained increased attention because it offers myriad advantages to a team of agents. While biological agents inspire robotic teams, and technology is advancing to allow swarms of inexpensive and interchangeable robots to be used in many fields, more and more frequently swarming behaviors are used to conceal the team’s goals. Although those collective behaviors are usually easy to program to a robotic swarm, solving the inverse problem of reconstructing and deciphering the high-level goals from the macroscopic description of the swarm remains a challenge. This project will focus on developing and testing novel computational methods to decipher and reconstruct high-level goals from the macroscopic analysis of collective behaviors in swarms using a cyber-human approach. State-of-the-art image processing tools will be aided by the responses of the human observer’s brain to the collective behaviors of the swarm. In other words, geometric description of the macroscopic view of the swarm facilitated by image-processing tools for crowd analysis, will be fused with principles of pattern recognition applied to the brain-activation of a human observer, to result in a human-machine symbiotic system that will be able to extract high-level behaviors and goals of a swarm. New computational and mathematical principles of cognition will be introduced in order to form a symbiosis between human and machine systems for comprehensive situation awareness, collaborative assumptions about adversarial agents, and shared decision making. The outcome of this project is a cyber-human system able to extract the centralized controllers and primitive behaviors of a swarm, and aid the human operators in deciphering and predicting goals and behaviors of an adversary swarm.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 30, 2018
- Source ID
- FA95501810221
Entities
People
- Panagiotis K Artemiadis
Organizations
- Air Force Office of Scientific Research
- Arizona State University
- United States Air Force