Synchronous Rendezvous for Heterogeneous Robotic Sensor Networks in Geophysical Flows
Abstract
Existing mobility strategies for robot data mules operating in a stationary sensor field rely on geometric optimization strategies [2, 3]. Therefore, they do not readily extend to non-stationary sensor network settings. If AUVs and ASVs are to move from one drifter node to another and assist in harvest, exchange, and upload data, new algorithms for motion planning and control are required. Specifically, the development of such an oceanic sensor network requires robots to achieve temporal synchronization on a set of moving rendezvous points given by the drifter positions. The objective of this work, therefore, is to develop distributed control strategies for autonomous vehicles to: 1. Achieve distributed synchronous rendezvous on a set of moving rendezvous points whose motions are dictated by surrounding flow field; 2. Leverage PI Hsieh s ONR YIP outcomes to develop energy-efficient motion control strategies for synchronous rendezvous by leveraging the dynamics of the fluidic environment; and 3. Experimentally validate the proposed strategies using the ONR funded research instrument: the multi-robot Coherent Structure Testbed (mCoSTe) developed by PI Hsieh as part of her ONR YIP efforts. The novelty of this work lies in the synthesis of ideas from nonlinear dynamical systems theory, transport theory, and robotics to develop a sensing and control framework for collaborative unmanned systems operating in dynamic and uncertain environments. Since drifter and vehicle motions are governed by dynamics of the geophysical flow, motion planning and control strategies for robotic sensor networks must account for their effects. The proposed approach will leverage the dynamics of the surrounding flow field to identify rendezvous locations. Synchronous rendezvous at the identified locations would then be achieved through distributed control. While the focus of this work is on marine vehicles, the proposed methods are general and can apply to any autonomous vehicle whose dynamics is tightly coupled with the environmental forces, e.g., a micro-aerial vehicle. As such, success of the proposed activities will improve the autonomy of various Naval platforms and directly affect the Navy s abilities to navigate the oceans, increase the energy-efficiency of existing robotic sensor networks, and provide greater situational awareness for Naval applications.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 03, 2016
- Source ID
- N000141612216
Entities
People
- Mong-ying Hsieh
Organizations
- Drexel University
- Office of Naval Research
- United States Navy