AVATAR - Automated Virtual Agent for Truth Assessments in Real-Time
University of Arizona
There are many circumstances, particularly in a border-crossing scenario, when credibility must be accurately assessed. At the same time, since people deceive for a variety of reasons, benign and nefarious, detecting deception and determining potential risk are extremely difficult. Using artificial intelligence and non-invasive sensor technologies, BORDERS has developed a screening system called the Automated Virtual Agent for Truth Assessments in Real-Time (AVATAR). The AVATAR is designed to flag suspicious or anomalous behavior that warrants further investigation by a trained human agent in the field. This screening technology may be useful at Land Ports of Entry, airports, detention centers, visa processing, asylum requests, and personnel screening.
The AVATAR has the potential to greatly assist DHS by serving as a force multiplier that frees personnel to focus on other mission-critical tasks, and provides more accurate decision support and risk assessment. This can be accomplished by automating interviews and document/biometric collection, and delivering real-time multi-sensor credibility assessments in a screening environment. In previous years, we have focused on conducting the basic research on reliably analyzing human behavior for deceptive cues, better understanding the DHS operational environment, and developing and testing a prototype system.
New Immigrant Survey
University of Arizona
Little is known about the origins of legal immigrants, other than their countries of birth or last residence. We currently know little about their pre-immigration labor market experiences and the ability of immigrants to translate those experiences into labor market success in this country. A key to a more coherent picture of the American migration experience is knowing why immigrants choose to migrate to the U.S. and the factors that make this decision permanent or temporary. We also need a better understanding of the factors affecting the assimilation of immigrants and their children.
This project enlists leading and promising researchers to utilize the New Immigration Survey (NIS) data in statistical and quantitative analyses of immigration. Awards will be given based on the innovativeness and quality of the proposed research and use of NIS data, including faculty awards of $30,000 each and young researcher awards (postdoctoral fellows or doctoral students) awards of $12,000 each.
Biometric Identification: Research Directions
West Virginia University
The past two decades has seen a substantial increase in biometrics activity accompanied by the deployment of biometric systems in diverse applications ranging from laptop access to border control systems. The inclusion of biometric evidence in military and criminal courts necessitates a careful examination of the scientific basis for biometric recognition. In particular, there is an urgent need to systematically review the scientific literature to determine if some of the common assumptions made about biometric traits with respect to criteria such as universality, uniqueness, permanence, measurability, performance, acceptability and circumvention, is borne out in the academic literature. Thus, the purpose of this study is to:
(a) Identify gaps in existing research and the implications on operational system risks; and
(b) Provide recommendations for further research and deployment scenarios. Pupil Dilation Characterization and Mitigation Studies
Border Patrol Checkpoint Effectiveness: Models and Metrics
University of Arizona
The Checkpoint Study project has been initiated to help Customs and Border Protection - Office of Border Patrol (CBP-OBP) assess the effectiveness of traffic checkpoint operations for the public good. The key goals of the project are to evaluate and address 1) checkpoint data integrity, consistency and accuracy, 2) measures of checkpoint impacts on local communities and 3) effectiveness metrics and models, as pointed out in the GAO Report No. GAO-09-824.
Localization and Tracking of Vehicles, Cargo and Persons
University of Minnesota, Twin Cities Campus
The overall objective of the work in this project is to develop solutions for accurate, cost effective and reliable tracking and localization of designated vehicles and cargo entering the country via the borders. The solutions developed use (to the maximum extent possible) existing technologies and are therefore cost effective. The solution developed for vehicle tracking leverages the wide-acceptance of GPS as a solution for fleet tracking. It adds to this advanced signal processing techniques to ensure that designated vehicles cannot spoof or deceive tracking efforts, thereby, providing assurance that a vehicle has (or has not) traveled along a designated route. The cargo tracking solution does the same for individual pieces of cargo (separate from the vehicle) that may be moved from one vehicle to another in the process of transit.
An Assessment of CBP Consequence Delivery Programs
Migration Policy Institute
Consequence delivery programs (including Expedited Removal, Operation Streamline, Mexican Interior Repatriation Program, Alien Transfer Exit Program, and Operation against Smugglers Initiative on Safety and Security) have been adopted across the Southwest Border, though most extensively in the Tucson Sector. Recent declines in apprehensions suggest these programs may be among the factors deterring repeat illegal entries. We will assess the effectiveness of various consequence programs by reviewing the literature, meeting with technical and policy experts, visiting the Tucson Sector to meet with Border patrol staff, and analyzing CBP IDENT data to identify recidivism rates (i.e., patterns of repeat illegal entry attempts). Following the site visit and data analysis, we will discuss findings at an invitation-only roundtable with technical and policy experts similar to those assembled for our previous border metrics project. We will produce a report summarizing our findings and the roundtable discussions, and drawing conclusions about the deterrent impacts and overall effectiveness of consequence delivery programs.
Determinants of DACA Applications: A Multi-level, Bi-national Study of Incentives, Deterrents, and Consequences of Decisions to Seek Deferred Action for Childhood Arrivals
University of California - San Diego
This project will identify and estimate factors affecting the rate of applications to the Deferred Action for Childhood Arrivals (DACA) program. Our core research question is how immigrant communities in the United States and Mexico are interpreting and responding to DACA, and how these responses determine who applies for the program and who does not. We will study determinants of application decisions that operate at both the individual and receiving-community level. We will also document the short-term consequences of participation in the DACA program on immigrant integration. Our methodological approach to these questions will combine field surveys in southern California and Oaxaca, Mexico with analysis of individual-level data on DACA applicants obtained via Freedom of Information Act (FOIA) requests. The aim of the research is to provide guidance for improving program administration and increasing participation among DACA-eligible immigrants.
Coverage Estimates of the U.S. Mexican-Born Population and Measures of Unauthorized Migration
University of California-Irvine
Information about the extent to which the Mexican born are captured in U.S. Census surveys directly feeds into estimates of the total unauthorized Mexican population, yet little is known about their coverage error. Our earlier BORDERS-supported research used two methods (the net migration and death registration methods) to produce coverage error estimates for 1995 through 2005. We now plan to update, extend, validate, and extrapolate these results. We have four objectives: (1) to update the earlier estimates using the death and migration methods through 2010, (2) to extend the research by adding estimates using a third method, a birth registration method, for the entire 1995-2010 period, (3) to validate the results by conducting robustness checks on the results, and (4) to extrapolate the findings on Mexican-born coverage in order to assess of the adequacy of the levels of undercount used in current estimates of unauthorized Mexican migration.
Development and Testing of a Cell-Phone Signal Based GPS-Denied Navigation System for Small UAVs
University of Minnesota
A prototype Small UAV (SUAV) navigation system capable of generating a Position, Navigation and Time (PNT) solution in GPS-denied, urban operational environments will be developed. The system generates a PNT solution by relying on a Dead Reckoning (DR) navigator as a core (or inner-loop) sub-system aided by periodic position fixes from cell-phone signals. The DR system uses sensors that are typically found on SUAV autopilots (namely, an Inertial Measurement Unit or IMU; a magnetometer triad; and an air speed sensor or pitot tube) to generate a high bandwidth solution of position, velocity and attitude. To mitigate DR system drift, cell-phone signals will be used to periodically generate a position fix (and a timing solution) which is fused with the DR system solution via an Extended Kalman Filter (EKF). The cell-phone position fixes are generated by a machine learning code which determines the cell-phone signal “finger print” of a given area.
Pupil Dilation Characterization and Mitigation Studies
This study will focus on the design of a mathematical model for characterizing the dynamics of the pupil and the iris of the human eye in response to changes in ambient light. In particular, the study will establish if existing iris normalization models, used by several commercial iris recognition systems, have to be modified in order to account for the non-linear geometric deformations observed in the iris texture in response to changes in external illumination. This study is expected to impact the design of iris recognition systems and is, therefore, critical to Homeland Security.
The work will be undertaken in 4 different tasks:
- Task 1 - Modeling Pupil Deformation
- Task 2 – Data Collection
- Task 3 – Model Evaluation
- Task 4 – Dilation Mitigation Model