Real-Time Perception and Planning in Network of Agents

Abstract

This project proposes a rather unified framework for several fundamental problems in sensing, perception, and control of multi-agent systems. We propose that dynamical spacetime sampling of streaming data, its representation for perception, nonlinear behavior of agents, and task planning formulations involving onlinearities can be formulated in a unified manner via frame theory. If successful, the outcomes of this project will provide a map on how to evaluate robustness and performance of closed-loop systems subject to streaming data. The theoretical findings of this project will be implemented to achieve vision-based swarm of multiple quadcopters.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 20, 2019
Source ID
N000141912478

Entities

People

  • Nader Motee

Organizations

  • Lehigh University
  • Office of Naval Research
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Vision.
  • Control Systems Engineering.