Project Number:
09-17
Research Project:
Risks and Recoveries from Extreme Disruptions in Freight Transportation System in a Megacity: Case Study for the Greater Los Angeles Area
P.I. Name & Address:
Suya You
University of Southern California
Department of Computer Science
Los Angeles, CA 90089
Email: suyay@imsc.usc.edu
Website: http://graphics.usc.edu/~suyay/
Phone: (213) 740-4495
Fax: (213) 740-5807
Co-P.I.
Ulrich Neumann
University of Southern California
Department of Computer Science
Los Angeles, CA 90089
Email: uneuman@graphics.usc.edu
Website: http://graphics.usc.edu/cgit/un.html
Phone: (213) 740-4489
Fax: (213) 740-5807
Project Objective:
Creating and maintaining an accurate and up-to-date road infrastructure database is crucial to many transportation applications including transportation infrastructure management, traffic situational awareness, safety analysis, and mission planning and tactical decision-making for incident and emergency responses. Nowadays, Intelligent Transportation Systems (ITS) continuously gain ground in transportation management. A digital topographic database is an essential part of the ITS, which requires accurate, high-density spatial models of road infrastructures. In addition, accurate road maps and databases are high public demand for travel planning, route guidance, and real-time travel navigation.
A complete road network model consists of road segments and intersections that joint the road sections. In particularly, accurately identifying and labeling of the road intersections in road networks is important, especially for safety management. Intersection safety has become a serious problem in the United States. FHWA has been initiating a variety of researches and strategies focus on improving intersection safety, including to developing a comprehensive database of road network and intersections for major US roadways. Many states including California also started to conduct similar efforts and strategies.
However, current road infrastructure databases and the methods to produce such as databases are insufficient to meet these needs in terms of accuracy, confidence, completeness, and automation. Many existing digital maps and road databases are still generated from the old paper based topographical maps, which may contain significant spatial errors. Accurate road databases have not yet existed for vast areas, particularly in areas with rapid expansion. Many existing databases need to be updated to capture new road condition and expansion.
The research goal of this proposal is to pursue a step change in the approach and technique for improving and extending the capabilities of creating, modeling and maintaining accurate and up-to-date road infrastructure models and databases for transportation managements and services. Our research efforts are to assess, define, and use the unique spatial and spectral characteristics of the new, advanced sensor techniques from aerial imagery and LiDAR for automated road extraction and road quality mapping.
Leveraging many years of research in geospatial data processing and modeling, we propose to integrate and extend our capabilities to include automated extraction, modeling, and labeling of road features and attributes from multiple sensor data sets. A number of theoretical and experimental studies lead us to pursue an innovative approach that merges the power of perceptual grouping and classification with sensor cues, geometric invariants, and machine learning under a unified framework to tackle these problems. This new approach has the potential for automating the extraction and mapping of complex road networks from data acquired from varied sensor sources. In addition, the same process also allows for a constrained optimal estimation of various terrain features and attributes including edges, curves, junctions, and surfaces and their relationships, thereby producing hierarchical data representations under a consistent framework. Most important, we believe that the process of labeling the model elements as buildings, vegetation, roads, and intersections will be possible within this framework. We anticipate a significant step reduction in the human time and effort required to produce and update accurate road maps and models to Caltrans' infrastructure databases. We believe that success with this research will have a high impact on current and future intelligent transportation systems and applications.
Task Descriptions
Task 1: Technical survey, user case identification and requirement determination
a. Establish research team and activities, including hire student researchers.
b. Conduct surveys of available technologies and capabilities, and determining METRANS requirements and relevant technologies essential for this research.
c. Contact METRANS and Caltrans personnel to determine use-cases and ascertain their views and knowledge. We intend to identify a use-case from a dense urban area such as Los Angeles or San Francisco area for case study.
Task 2: Data collection and core technology developments
a. Collect aerial image and LiDAR data for the designated sites. Currently we already have several data sets including portion of LA downtown and entire San Francisco area. In addition, we also have collaborative relationship with several geospatial service companies (Airborne1 Inc, Sanborn) for data collection.
b. Develop algorithms for rapid extraction of road network and intersections from imagery and LiDAR. We will use our existing framework as platform for new algorithm developments and system demonstrations since it already has examples of many capabilities and we can quickly extend and prototype new functionalities.
c. We will also consider an optional integration with existing/planned Caltrans databases, providing enhanced and up-to-date road maps and attributes.
Task 3: Performance evaluation, demonstration and reporting
a. We will conduct demonstration to METRANS and related agencies to demonstrate the developed technology and its benefit to METRANS. The new findings resulting from the research will be documented and submitted to peer-referred journal and conference. Quarterly progress reports and final report will be prepared and submitted to METRANS.
Milestones, Dates:
August 15, 2008 – September 14, 2009
Total Budget:
$90,000
Student Involvement:
One graduate student at 50% effort, 9 months
One graduate student at 25% effort, 9 months
Relationship to Other Research Projects:
Related to 03-23; part of the infrastructure focus area
Technology Transfer Activities:
Project report will be posted soon
Potential Benefits of the Project:
This research will reduce the time and effort required to produce and update accurate road maps and models to Caltrans infrastructure databases and will have a positive impact on current and future intelligent transportation systems and applications.
TRB Keywords:
Intelligent transportation systems (ITS), LiDAR, infrastructure, Caltrans
