Development and Demonstration of Enhanced Transient Flocking Behaviors for a Large Group of Mobile Agents
Abstract
Research Summary: Development and Demonstration of Enhanced Transient Flocking Behaviors for a Large Group of Mobile AgentsPI: Yan Chen, Arizona State UniversityMulti-agent systems, especially consisting of mobile agents, are extensively studied due to their great potentials on commercial and military applications. One notable example is that autonomous underwater vehicles (AUVs), as well as unmanned aerial vehicles (UAVs) and connected and automated vehicles (CAVs), are applied in environment exploration, data collection, and tactical surveillance. Regardless of specific applications of different multi-agent systems, one main purpose is to achieve desired and networked collective motions through appropriate control. Many challenges actually need to be overcome for the control purpose of multi-agent systems. Although great efforts were taken to tackle (swarm or flocking) motion control issues of multi-agent systems in the literature, some key challenges still need to be further explored. For example, 1) multi-agent systems performing tasksin unstructured, time-varying, and adversarial environment often requires advanced transient flocking motions, such as fast-convergence, energy-efficient, and agile flocking behaviors; 2) balancing or compromising strategies are needed to coordinate different, even conflicting, transient flocking behaviors; 3) experimental demonstrations of flocking control with advanced transient motions on a large group of (e.g., more than 50) low-cost mobile agents are still missing.The overarching goal of this proposal is to develop asystematic and fundamental framework that can achieve advanced transient flocking motions for multi-agent mobile systems, and to experimentally demonstrate the theoretical development and performance improvement with respect to existing baseline flocking strategies.To achieve this goal, three major novel research elements will be investigated. 1) Novel transient flocking control theories willbe developed to control and optimize collective motions of multi-agent mobile systems. Specifically, fast-convergent, energy-efficient, and agile flocking with respect to permanent obstacles (constraints) will be respectively explored, which are absent in existing flocking theories. 2) To integrate the proposed transient flocking control methods together, balancing and compromising strategieswill be created based on a hierarchical framework to integrate the high-level flocking control (based on simple point-mass models) with the low-level tracking control (based on nonlinear agent models). 3) Experimental validations of the proposed transient flocking theories and the balancing strategies will be completed based on a large group of low-cost homogeneous CAVs (e.g., more than 50) and compared with the existing baseline flocking control, in order to extend the main research results to other automated multi-agentsystems, such as AUVs and UAVs.The summary is approved for public release.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- May 05, 2021
- Source ID
- N000142112403
Entities
People
- Yan Chen
Organizations
- Arizona State University
- Office of Naval Research
- United States Navy