You are here

A Resource-aware metadata-based Information Sharing: Achieving Scalability and VoI in future autonomous networks

Description:

TECHNOLOGY AREA(S): Info Systems 

OBJECTIVE: The objective of this topic is to develop resource efficient methods and techniques that generate and annotate metadata based on information that has been retrieved from Army tactical networks that deploy artificial autonomous agents. The goal is to improve the accuracy of information queries, with this accuracy determined by quantitative criteria that reflect the risk in misidentifying what information is relevant for the Army mission at hand. 

DESCRIPTION: The Army vision of artificial autonomous agents collaborating with mounted and dismounted forces in order to perform a wide range of mission operations will require scalable and robust networking solutions. Artificial agents of different types and complexity will consequently form a heterogeneous network that has variable resources and capabilities, and will need to coordinate and interact with each other to allow the completion of the required mission tasks while respecting the limited resources available in a tactical network. Due to the limitations of communication bandwidth, storage and processing capabilities of tactical edge networks, it is impossible to disseminate all the generated information (e.g. images and videos) to the agents that need it. For example, autonomous aerial drones with mounted cameras can generate images to aid in mission planning by uploading all the images/videos they record to a server, and the server will then utilize content-based techniques to resolve user queries over the images and videos. However this approach can utilize an appreciable amount of network bandwidth, and the storing and processing of images and videos can utilize a significant amount of disk space and processing power on the server. But realistically only a small percentage of the uploaded images or videos may actually be useful to any participating agent in the network. Therefore the resources utilized to upload, store, and process the remaining images and videos will be wasted. These considerations call for the design of an information access methodology on the network that will be aware of network, computational, and relevance constraints. Such a methodology will significantly enhance the network’s ability to transfer relevant information, and will therefore increase the likelihood of mission effectiveness. To this end, this methodology should follow some of the current techniques in information science, in particular the current paradigm of neural networks and deep learning in order to generate metadata from the data acquired by the agents in the network. The goal will be to develop techniques that will: (i) Generate/annotate metadata based on embedded sensors of the autonomous agents (both artificial and human) that will optimize the network resource utilization and processing power of the agents using algorithms that will scale appropriately. (ii) Assist human users in annotating metadata in order to increase query accuracy; (iii) Select and retrieve images and videos based on the available network resources and what is called the Value of Information (VoI), a concept that captures quantitatively the relevance of the information to mission tasks and completion. The research should address what strategies for generating and annotating metadata will increase the accuracy of matched queries and tasks for the given mission. This includes auto-generated data from artificial autonomous agents as well as data annotated by the human user. The research should also address information retrieval techniques that incorporate strategies to select advantageous combinations of modalities (e.g. video, text, images) that can significantly increase the query accuracy, the VoI, while still being aware of network availability and utilization. The network that deploys this methodology is expected to operate in contested and congested environments with intermittent communication links, and therefore the agents will need to take advantage of all short-lived, high-rate communication opportunities if and when they arise. 

PHASE I: Explore and design strategies and algorithms that annotate, organize and select metadata and information content queries while being aware of the network conditions and the value of information requirements. Define a framework for intelligent capture of the interactions between human and artificial autonomous agents. Use this framework to share information over wireless heterogeneous network. Demonstrate viability of solution through modeling and simulation. 

PHASE II: Develop specification and software implementation of the proposed algorithms and techniques from Phase I. Demonstration of the scalability properties of the proposed solution using a combination of artificial autonomous agents and human agents in wireless mobile network in combination with emulated network. Demonstrate the capabilities using a network of wireless mobile nodes under a relevant outdoor scenario. 

PHASE III: This research can enhance network capability for supporting intelligence gathering of information in different coalition networks. Users (artificial and human) in these network settings are likely to generate large volumes of content consisting of images/videos. With the metadata based information access, we can significantly enhance the information carrying ability of the tactical network, and then lead to better success in missions. In addition to military applications, manned and unmanned teaming efforts within First Responders and Homeland Security are expected to grow and benefit from the metadata based information access. Envisioned improvements to be provided by this topic in terms of network efficiency and scalability can also be inserted in these applications and thus enable broader use of their capabilities. 

REFERENCES: 

1: T. Dao, A. Roy-Chowdhury, H. Madhyastha, S. Krishnamurthy, T. La Porta, "Managing redundant content in bandwidth constrained wireless networks," In ACM International Conference on emerging Networking Experiments and Technologies, 2014.

2:  Richard E. Mayer., Cognitive theory of multimedia learning. In the Cambridge Handbook of Multimedia Learning, pp. 43-71. Ed. Richard E. Mayer. New York, NY: Cambridge University Press, 2014

3:  Richard E. Mayer. The promise of multimedia learning: using the same instructional design methods across different media. Learning and Instruction 13(2):125-139. 2003.

4:  Rebecca J. Passonneau, Emily Chen, Weiwei Guo, Dolores Perin. Automated Pyramid Scoring of Summaries using Distributional Semantics. Proceedings of the 2013 Annual Meeting of the Association for Computational Linguistics. 2013.

5:  Qian Yang, Rebecca J. Passonneau, Gerard deMelo. PEAK: Pyramid Evaluation via Automated Knowledge Extraction. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, 2016

6:  M. Uddin, H. Wang, F. Saremi, G. Qi, T. Abdelzaher, and T. Huang, "PhotoNet: A similarity-aware picture delivery service for situation awareness," in IEEE Real-Time System Symposium, 2011.

7:  Yi Wang, Wenjie Hu, Yibo Wu, and Guohong Cao, "SmartPhoto: A Resource- Aware Crowdsourcing Approach for Image Sensing with Smartphones," ACM Mobihoc, 2014

8:  Y. Wu, Y. Wang, W. Hu, X. Zhang, and G. Cao, "Resource-Aware Photo Crowdsourcing Through Disruption Tolerant Networks," IEEE International Conference on Distributed Computing Systems (ICDCS), 2016.

KEYWORDS: MUM-T, Metadata, Artificial Agents, Information Network, Communication Network, Routing, Wireless Network, Drone, SUAV 

CONTACT(S): 

Mitesh Patel 

(443) 395-7630 

mitesh.p.patel.civ@mail.mil 

Bart Panettieri 

(443) 395-7371 

US Flag An Official Website of the United States Government