| NSF Org: |
CNS Division Of Computer and Network Systems |
| Recipient: |
|
| Initial Amendment Date: | September 14, 2015 |
| Latest Amendment Date: | September 14, 2015 |
| Award Number: | 1527287 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Marilyn McClure
mmcclure@nsf.gov (703)292-5197 CNS Division Of Computer and Network Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2015 |
| End Date: | September 30, 2019 (Estimated) |
| Total Intended Award Amount: | $125,001.00 |
| Total Awarded Amount to Date: | $125,001.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
110 8TH ST TROY NY US 12180-3590 (518)276-6000 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
110 8th Street Troy NY US 12180-3590 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | CSR-Computer Systems Research |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
Many of today's vehicles are equipped with GPS, networking capabilities, computational resources, and a variety of environment and vehicle performance sensors. The combination of these technologies and the growing presence of such vehicles offer a powerful platform for large-scale road sensing applications to assist drivers in route planning and safe driving. As the number of participating vehicle grows, the sheer amount of data and the cost of its transmission in cellular networks make current, centralized solutions unsustainable. This project proposes an integrated approach to overcoming algorithmic and systems barriers to the cost and scalability challenges of collaborative road sensing. The project conducts four main research tasks: (1) obtain correlated measurements of vehicle sensor readings and network connectivity, (2) reduce the cost of vehicle-to-cloud (V2C) communication through vehicle-to-vehicle (V2V) message delegation, (3) develop new communication-efficient algorithmic approaches for collaborative road sensing, (4) evaluate the communication and algorithmic solutions in an emulation framework that pulls together vehicle sensor data, measurements of cellular and V2V connectivity, and traffic flow data from a I-90/94 corridor information system. Results of this project will include reliable connected vehicle communication architecture for both urban and rural settings and theoretically sound decentralized algorithms for sensing and estimation that will be of interest to the broader distributed computing systems community.
The communication architecture and sensing algorithms will serve as a prototype for large-scale collaborative, real-time road sensing. The widespread adoption of such an approach will provide drivers with better awareness of hazardous driving conditions, supporting safer and more efficient transportation. Techniques and systems developed in this project will also have applications in settings that require analysis of vast amount of sensed data that is distributed over a large number of devices or systems, for example, in collaborative airspace monitoring in airplanes and environment monitoring in the Internet of Things.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
The goal of this collaborative project was to develop a framework and algorithms for road monitoring using vehicle sensor networks in conjunction with cloud computing resources. The main research tasks were: (1) devise hardware and software to obtain data sets of measurements of vehicle sensor readings and network connectivity from real-world logging drives; (2) design algorithms for efficiently analyzing this kind of data using vehicle compute and communication resources (edge devices) where possible, and incorporating cloud computing resources when needed to generate accurate, real-time results; (3) evaluate the algorithmic solutions in a framework that pulls together sensor data, measurements of network connectivity, and realistic traffic and device density models.
The intellectual merit of the project includes the following contributions:
- We developed an Arduino-based sensor package, MOVESET, that is mounted on a vehicle. MOVESET collects sensor data on temperature and humidity, as well as measurements of network connectivity using cellular and DSRC radios. We used MOVESET on several logging drives in Montana and published the results in the Vehicular Networking Conference. The MOVESET design plans are available at https://nl.cs.montana.edu/moveset/.
- We created Spindle, a prototype Map-Reduce framework for mobile networks, which integrates with a cloud-based Apache Spark instance, to perform aggregation of vehicle sensor data in the hybrid vehicle-cloud platform. Spindle is available at https://github.com/rpi-nsl/NSL-Spindle.
- We devised graph-theoretic abstractions for computation over the hybrid network. Using these abstractions, we developed theoretical characterizations of algorithm performance and the relationship to the network topology and characteristics. This work resulted in several conference and journal publications.
- We developed resource allocation mechanisms for cloud platforms. These mechanisms dynamically determine the set of resources that is required for efficient real-time data processing depending on the input workload. It is crucial to allocate sufficient computing resources so that data streams from connected vehicles can be analyzed in real time. Using our approach, we are able to guarantee robust and responsive performance of cloud data analytics software while minimizing the cost to application owners. This work resulted in several conference publications.
- We executed performance studies of hybrid edge-cloud platforms using models and data obtained from benchmarks. This work was published in two papers in a Utility Cloud Computing companion workshop.
The broader impacts of this work extend to a wide range of current and emerging computing platforms, including edge computing, cloud computing, and cyber-physical systems.
Last Modified: 12/26/2019
Modified by: Stacy Patterson
Please report errors in award information by writing to: awardsearch@nsf.gov.
