Adaptive planning for improved search performance
Abstract
We propose a three-year project to address both fundamental and practical challenges in subsea search and survey. The overall goal is to develop decision algorithms for Navy- relevant missions that maximize percent clearance in a xed time. Our overall approach is two-fold: we seek a collection of tools that address speci c applications, but we also seek general principles that are valid for a broad set of applications. In all cases, we favor rigorously justi ed approaches and well-de ned measures of performance. We will address three classes of challenges: 1. Development and analysis of decision algorithms for detection and classi cation missions where one or more vehicles are equipped with a wide-area sensor capable of detecting objects of interest, while the same or other vehicles are equipped with a short-range sensor capable of object classi cation. Once objects are detected with a long-range sensor, they must be classi ed using a short-range sensor. The objective is to develop decision rules that can be implemented in real-time with only modest computational requirements, but that possess formal performance guarantees. 2. Development of methods to account for the cost of communication and coordination for missions where vehicles are not necessarily operating within communication range of each other, and thus the cost of communication is dominated by the cost of pausing a mission in order to transit toward and communicate with another vehicle. The objective is to determine when the expected bene t of communication outweighs the extraordinary cost of communication. 3. Rigorous and extensive evaluation of our results for mission-relevant scenarios using high- delity simulation tools being developed at NSWC Panama City. This task will be performed in collaboration with engineers and scientists at NSWC Panama City. The objective is to ensure that our fundamental results lead to useful algorithms that can be implement in Navy-relevant scenarios.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 10, 2016
- Source ID
- N000141612092
Entities
People
- Daniel J. Stilwell
Organizations
- Office of Naval Research
- United States Navy
- Virginia Tech