Nikolaos Papanikolopoulos - US grants

A dexterous manipulator with seven degrees of freedom shall be obtained for research projects involving mobile robots with high degrees of freedom. These projects include: * Behavior-based control. * Impedance algorithms. * Error recovery * Parallel numerical algorithms for redundant robots. * Robust and fault-tolerant robotics visual tracking and servoing.

This research pursues a model-based approach for visual tracking and servoing in robotics. Deformable active models are explored as an effective way for tracking a rigid or semi-rigid (possibly partially occluded) object in movement within the manipulator's workspace. Deformable models imitate, in real-time, the dynamic behavior of elastic structures. These computer-generated models are designed to capture the silhouette of rigid or semi-rigid objects with well-defined boundaries, in terms of image gradient. By means of an eye-in-hand robot arm configuration, the desired motion of the end-effector is computed with the objective of keeping the target's position and shape invariant with respect to the camera frame. Optimal estimation and control techniques (LQG regulator) techniques are used in order to deal with noisy measurements provided by the vision sensor. Preliminary experimental results show that the deformable models help to achieve robust tracking even when the target is partially occluded. These techniques will be applied to assembly tasks that involve a combination of vision and force robot control. Finally, the algorithms will be tested by applying them to transportation problems such as pedestrian tracking and vision based vehicle- following.

Flexibility and adaptability of a robotic manipulator can be achieved by incorporating vision and sensory information in the feedback loop. Previous research by the PI introduced a framework called ``controlled active vision'' for efficient integration of the vision sensor in the feedback loop. This framework emphasized eye-in-hand robotic systems (where the vision sensor is mounted on or close to the manipulator's end-effector) and was applied to the problem of robotic visual tracking and servoing with very promising results. This research extends the framework to other problems of eye-in-hand robotic systems such as the derivation of depth maps from controlled motion; the vision-guided, automatic grasping of moving objects; the active calibration of the robot-camera system; the problem of automatically detecting moving objects of interest; and the computation of the relative pose of the target from the camera. In addition, the new work investigates issues such as the stability and the robustness of the algorithms. All the work is experimentally verified on the Minnesota Robotic Visual Tracker (a flexible eye-in-hand robotic system). This research has potential applications to transportation (e.g., pedestrian detection and tracking, vision-based vehicle following), inspection, and assembly (e.g., vision-guided manipulation of moving objects).

9729875 Heimdahl, Mats, P.E. University of Minnesota, Twin Cities CISE Research Instrumentation: Applying Software Engineering Methodologies to Robotics Tasks: A Cross Disciplinary Approach This research instrumentation enables research projects in:- Rapid Prototyping and Synthesis of Software for Embedded Systems, - Studies of Different Issues in Vision-Based Robotic Grasping,- Sensor Fusion in Vehicle Applications, and - Behaviors of Multiple Self-Interested Robots. To support the aforementioned projects, the Department of Computer Science at the University of Minnesota will purchase a VXI-bus embedded real-time software development environment, several mobile platforms, and robot simulation software which will be dedicated to research in software engineering, robotics, and artificial intelligence. The equipment will be used for several research projects, including in particular: requirements-based prototyping and synthesis of control software for embedded systems, closed-loop vision-based robotic grasping, sensor fusion for vehicle applications, and behaviors of multiple self-interested robots. In addition to the contributions of the individual projects, the collaboration between the PIs will help evaluate cutting edge software engineering techniques on realistic projects (for example, evaluation of scalability, evaluation of ease of use, and applicability to the domain) as well as evaluate the use of rigorous software engineering techniques in robotics and AI applications (for example, the effect on reuse of control algorithms, and the effect on system performance, robustness, and safety). The equipment purchase includes (1) a complete VXI-bus embedded software development environment, (2) one VXI-bus embedded computer with interfaces for sensing and control applications, (3) low-cost mobile platforms serving as testbeds for the proposed projects, and (4) robotics simulations software to be used in task planning and software prototyping.

Papanikolopoulos
Institution: University of Minnesota - Twin Cities.
NSF/USDOT: Real-time collision Warning at Traffic Intersections

Collisions between vehicles at urban and rural intersections account for nearly a third of all reported crashes in the United States. This research entails developing some of the components of a real-time system which uses cameras to continuously gather traffic data at intersections (e.g., vehicle speeds, positions, trajectories, accelerations/decelerations, vehicle sizes, signal status, etc.), applies efficient algorithmic techniques to detect potential collisions and near-misses, and then issues suitable countermeasures. At this stage, the goal is to establish the feasibility of this approach using both computer simulations and field tests at actual intersections (urban intersections in the Twin Cities---Minneapolis and St. Paul, MN). The proposed work has as main emphasis the design of efficient and robust computer vision and collision-detection algorithms that can address the intersection collision warning problem.

EIA-0224363
Nikolaos Papanikolopoulos
Daniel L. Boley; William K. Durfeet; Maria L. Gini; Bradley J. Nelson
University of Minnesota-Twin Cities

CISE RR (Collaborative): Teams of Miniature Mobile Robots

This project, redesigning and manufacturing a team of Scout-Ranger robots, improvises on the current scout design through novel design schemes, software development, sensory fabrication, and processing sensory data in real time. The research extends work based on a previous generation of Scouts. The institution developed a heterogeneous robotic team emphasizing the "Scout" robot, launched and controlled by a larger platform, the "Ranger." After deployment, Scouts have a unique combination of mobility modes including rolling and hopping, multiple sensing capabilities for navigation (e.g., cameras, microphones), full communications for data and instructions (controlled by cellular phones), and onboard computational resources. Their functionality is increased by actuated wheels and miniature grappling hooks and can serve applications such as reconnaissance, earthquake rescue operations, homeland security space exploration, fire rescue missions, hostage release operations, etc. Their design requires a compromise in power, sensor types, locomotion, and size. However, rather than concentrating on size, this work focuses primarily on communications, sensor fabrication, and sensor placement. Solutions to research problems are sought in areas of miniature-robot design, communication for distributed robotics (especially in low-bandwidth situations), and resource allocation for distributed robotics. Specifically:
1. Designs of miniature robots: modifying original scout design, giving special care to image-processing
2. Software support for Robot Teams: sharing resources leading to decision making and planning
3. Tradeoffs between sharing communication channels and performance in motion detection
4. Sensor fabrication in the micro fabrication lab
5. Analysis of sensory data (images) of the team of robots; seeking optical positioning of the sensor to maximize coverage/redundancy
6. Analysis of traffic patterns (of humans) in a building
7. Detection of anomalies on those patterns, or detection of unusual situations (i.e. smoke): using method of Principal Direction Divisive Partitioning
8. Innovative uses of Scouts: tele-rehabilitation, dynamic hopping devices, etc.
The educational plan includes the incorporation and mentoring of women and minority students. The robots will be used not only in Computer Vision and Robotics classes, also, in Algorithms, Data Structures, Operating Systems, and Artificial Intelligence. Outreach activities, proposed for K-12, along with public outreach complement the research agendas.

This project investigates two problems associated with the monitoring of human activities: The first problem is tracking of articulated motion as a whole without identifying individual limb motion. The goal is to address certain shortcomings in previous solutions to this problem, the main shortcoming being their over-constrained nature. The proposed solution, which is presented as a real-time human tracking
system, will be capable of working under many difficult circumstances. The second problem is recognition of
articulated motion. The goal here is to show that the recovery of three-dimensional properties of the object or even two-dimensional tracking of the object parts are not necessary steps that must precede action recognition. The proposed approach uses motion features only. Unlike other similar approaches, the motion
features will be used in such a way to represent complex and long actions as well as to distinguish different actions with many similarities. Each action is represented as a manifold in the lower dimension space and matching is done by comparing these manifolds. As part of a homeland security scenario, its is planned to use these methods to monitor outdoor human activities based on the ability to recognize, for example, that a human runs in the opposite direction that a crowd moves.

This award supports the planning a multi-university, multi-disciplinary Industry/University Cooperative Research Center (I/UCRC) for Safety, Security and Rescue Robots (C-SSRR). C-SSRR will bring together industry, academe, and public sector users together to provide integrative robotics and artificial intelligence solutions for activities conducted by the police, FBI, FEMA, firefighters, transportation safety officials, and emergency responders to mass casualty-related activities. The need for SSRR has accelerated in the aftermath of 9/11 and a new research community is forming, as witnessed by the first IEEE Workshop on Safety, Security and Rescue Robotics in February 2003.

The Center will be built upon the knowledge and expertise of multi-disciplinary researchers in computer science, engineering, industrial organization, psychology, public health, and marine sciences at the University of South Florida and the University of Minnesota. Together, the two institutions support a research program in control of vehicles, human-robot interaction, and sensors and sensor fusion combined with rapid prototyping capabilities and access to users and high-fidelity testing sites throughout the country.

This project, a collaborative with 03-24977 Kostas Daniilidis at University of Pennsylvania and 03-25017 Joel Burdick at California Institute of Technology, addresses research issues key to an important application of robot teams and information technology (emergency response in hazardous environments for various tasks). The research sets 6 goals:
Development of new algorithms that enable collaborative sensing.
Development of distributed localization/mapping methods that leverage capabilities of the heterogeneous robots.
In-depth study of communication issues with emphasis on transparent integration of ultra wideband communication methodologies.
Development of methods for team coordination and dynamic distribution of tasks to robots.
Creation of algorithms for the presentation of sensory information to users.
Experimental validation of the scalability of the aforementioned algorithms and techniques.
The PIs use the Scout and MegaScout robotic platforms designed at the University of Minnesota along with other testbeds at CalTech and U Penn to conduct the research. The project integrates the algorithms with first responder teams, emphasizing realistic scenarios; mentors students from underrepresented groups in order to retain them in CS/EE programs; conducts outreach activities through demonstrations at local schools and youth groups; conducts workshops that emphasize cross-disciplinary interaction; creates web resources; innovates classroom uses of multi-robot teams; and includes parts of the research in design projects for seniors. The project also includes international collaboration with groups at NTUA (Greece) and the University Louis Pasteur-Strasbourg I (France).

This multi-university Industry/University Cooperative Research Center for Safety, Security and Rescue Research located at the University of South Florida and the University of Minnesota will bring together industry, academe, and public sector users together to provide integrative robotics and artificial intelligence solutions in robotics for activities conducted by the police, FBI, FEMA, firefighting, transportation safety, and emergency response to mass casuality-related activities. The need for safety, security, and rescue technologies has accelerated in the aftermath of 9/11 and a new research community is forming, as witnessed by the first IEEE Workshop on Safety, Security and Rescue Robotics in February 2003.

The Center will be built upon the knowledge and expertise of multi-disciplinary researchers in computer science, engineering, industrial organization, psychology, public health, and marine sciences at the University of South Florida (the lead institution) and the University of Minnesota.

Funds provided by the U.S. Army Research Laboratory will fund a project, "Algorithms and Software/Hardware Infrastructure for Distributed Miniature Robots", at the University of Minnesota research site of the Industry/University Cooperative Research Center (I/UCRC) for Safety Security and Rescue. The project pursues research into a hardware/software infrastructure to improve robotic response capabilities in the case of robot teams. The project builds on the current scout robot research and proposes some innovative robot designs that depart from the shape and functionality constraints of the initial platform.

Proposal #: CNS 07-07939 07-08344
PI(s): Isler, Ibrahim V. Papanikolopoulos, Nikolaos
Drineas, Petros; Trinkle, Jeffrey He, Tian
Institution: Rensselaer Polytechnic Institute University of Minnesota - Twin Cities
Troy, NY 12180-3522 Minneapolis, MN 55455-5200
Title: IAD: Collab Rsch: Research/Education Infrastructure Based on Modular Miniature Robot Systems

Project Proposed:
This collaborative project, developing a modular hardware and software infrastructure (miniature robots with the pertinent software), aims to identify minimum capabilities required for accomplishing tasks which are crucial for most robotics-sensor applications. The work involves investigating the relationship between the capabilities of individual units and the collective capabilities of the entire robotic team. The major innovation lies in the creation of a flexible design spectrum for both research and education based on the capabilities of the individual units. On the lower end of the spectrum, the basic design ""Explorer,"" a robot based on a Gumstix unit and earlier robot designs from one of the institutions, provides an inexpensive and yet significantly powerful solution. On the higher end, the platform ""MicroVision,"" a robot based on the Intel Pentium M processor and RoboAudioStix board, provides real-time video streaming and processing using standard off-the-shelf hardware and open-source video algorithms. The modularity of both designs allows the addition of special hardware capabilities. The infrastructure aims at low-cost, easy-to-use platforms that support research and can be adopted into courses. Specifically, this development enables design of robotics sensor-network systems that can
. Communicate reliably and efficiently via multi-hop in the presence of frequent network topology changes;
. Identify relative positions among members of a robotic team promptly and accurately for effective coordination;
. Enable energy-efficient marsupial maneuvering;
. Provide collaborative sensing with spatial, temporal, and spatiotemporal coverage guarantees; and
. Interact with environments through grasping and manipulation.

Broader Impacts: This infrastructure will be used for senior theses and K-12 students will be engaged via school visits and science field days. An annual Robot Camp will be held. The activities will be helped by the National Center for Engineering and Technology Education (NCETE). Coordination with Berea and Smith College, as well as UVA ascertains exchange among undergraduate- and minority-serving institutions.

This cross disciplinary project for instrument development combines the fields of computer vision (and computer science overall) with radiation therapy and behavioral sciences. This project is a joint venture among the Medical School, the Psychology Department, and the Institute of Technology. The goal is to develop a minimally invasive, full-body, patient tracking system to be used with the helical tomotherapy system currently in use at the institution. The purpose of this effort is to detect when a patient becomes misaligned during radiation treatment. Adding visual feedback to the helical tomotherapy treatment should improve the safety of the patient, reduce the considerable treatment time, and improve the cure rate. It should be noted that the proposed instrument has a wide variety of implications for other forms of radiation therapy and medical systems that require a specific patient pose with respect to the medical device. In order for these devices to revolutionize the treatment field, the interaction of the patient with the device should be studied from a behavioral sciences point of view. In this project, behavioral sciences will play an important role in capturing the behavioral patterns exhibited when the patient is in treatment, and in devising new patient placement protocols with profound impacts on cancer treatment.

During the early stages of this development, the proposed instrument will not be used in human trials, focusing instead on the following tasks:

- Development of efficient computer vision algorithms (structured light) for detecting the patient's movements in a non-intrusive way that simultaneously offers the possibility for creating a closed-loop helical tomography instrument,
- Selection of features, so cumbersome reflective markers are no longer needed,
- Creation of the probabilistic framework necessary to utilize the features found from multiple calibrated cameras to determine the probability distribution of likely body positions,
- Incorporation of behavioral science research that utilizes cyber-enabled principles to create patient-friendly medical devices and treatments, and
- Experimental validation of the proposed instrument and improvement of the instrument with cyber-enabled behavioral science input.

Broader Impacts:

This project introduces computer vision methodologies/hardware and cyber-enabled behavioral science research to radiation therapy with the objective of having safer and more effective radiation treatment. It may cause fundamental changes in the way that humans and medical devices interact. Educational programs will enable training to the instrument/methodologies. Some of these methodologies will be included in the curriculum. A website will be developed for dissemination of findings. Users from around the world will be given access. Outreach programs involving demonstrations to pertinent groups will be created.

0934327 University of Minnesota (UMN); Nikolas Papanikolopoulos
0934413 University of Denver (UD); Richard Voyles

The purpose of this proposal is to renew and expand the Center for Center for Safety, Security and Rescue Research (SSR-RC) as an NSF Industry/University Cooperative Research Center. This proposal is based upon UMN's successful completion of five years of operation of the SSR-RC; and the commitment by companies to join a research site at the University of Denver. UMN will be the lead research site for SSR-RC with the University of Pennsylvania (joined the Center a few years ago) and the University of Denver as research partners.

This proposal covers the renewal for the second-five years of UMN and the expansion to include UD. The proposed Center will provide integrative robotics, sensing, and artificial intelligence solutions in robotics for activities conducted by the police, FBI, FEMA, transportation safety, and emergency response to mass casualty-related events. The Center is built upon the knowledge and expertise of multi-disciplinary researchers in computer science, engineering, human factors, and psychology at the three institutions. The renewed and expanded Center will be successful because it builds on existing strengths developed during the first five years of operation. The Center will also educate and train researchers for industry and government.

The broader impact of the proposed center is to radically improve homeland defense in all dimensions. The proposed Center will encourage collaboration, and will nurture an emerging field of research and the associated industries, thus helping to establish the challenges of the field and acceptable research and evaluation methodologies. SSR-RC will expose students and faculty to state-of-the-art research projects of value to the industry, and plans to attract large companies to the SSR domains and energize innovative start-up companies. Students will have opportunities for industrial internships with members. Faculty in the SSR-RC will continue to aggressively recruit women and minority graduate students through the various I/UCRC supplemental programs, and to host annual summer camps for middle-schoolers from under-represented groups.

Abstract
Proposal #: 10-39741
PI(s): Papanikolopoulos, Nikolaos; Lim, Kelvin; Guillermo Sapiro
Institution: University of Minnesota
Title: MRI/Dev.: Video-Based Robotic Instrument for Behavioral Analysis and Diagnosis of At-Risk Children

Project Proposed:
The proposed set of tools constitutes a video-based robotic instrument which targets the domain of early diagnosis for children at risk of developing psychiatric disorders. As such, this proposal is at the disciplinary boundaries between computer science, psychology and psychiatry, and medicine. Proposed is the development of a robotic instrument that could observe and automatically analyze abnormalities in children, thus introducing a novel technology which can help identifying children at risk. Specific activities include:
- Development and clinical verification of instrumentation and clinical protocols to quantify mental disorders in children;
- Development and usage of computer vision and machine learning methodologies in the instrument;
- Development of statistical models to evaluate the available related data sets;
- Usage of a wide array of passive and active sensors and state-of-the-art 3D camera systems to collect and analyze the monitored data;
- Usage of robots and robot pets as a means to detect and treat mental disorders; and,
- Practical validation of the instrument at the Medical School.

Broader Impacts:
The recent usage of computer vision methodologies/hardware and robotics for detection of mental disorders in children, in itself, constitutes strong broader impacts. Planned are also educational programs (workshops, tutorials, etc.) that will enable training gathering of physicians and psychologists to the aforementioned methods/procedures, which would otherwise not be possible. Moreover, significant planned curriculum development at the participating institutions revolves around the instrument. In addition, outreach activities for middle-school students from underrepresented groups will take place, and so will outreach to various pertinent patient groups. This truly interdisciplinary project also plans to include international partners.

This project will develop algorithms to assist with the early diagnosis of children who are at risk of developing behavioral disorders. Previous research has indicated that two critical areas of behavioral investigation for use in identifying at-risk children have been abnormalities in motor activities and emotional range displays, especially of the face. Motor abnormalities are based on the observation that motor control involves the circuits of the brain associated with dopamine; these are also implicated in behavioral disorders. Many different disorders share the observation of disruption in the emotional range regulation, so facial expressions are included in the study.

To date, assessments of motor and emotional range have been done by the experts who view and rate videos of an individual. However, these expert, subjective ratings limit the analysis of behavioral conditions to only a narrow range of behaviors, work only for small populations of individual subjects, and are both costly and dependent on the observer's particular expertise. In order to enable wider population screening, automation is required. Innovative ways of capturing and quantifying the expertise of experts will be accompanied by metrics for assessing the evolution of the behavior. In addition, new computational tools will support evaluation of the effectiveness of interventions.

The broader impacts of the proposed work will involve improved mental health levels across the populations by providing a systematic approach for enhancing early detection, prevention, or mitigation of behavioral disorders and likely reduce the long-term costs of missed or late diagnosis. The research results will be blended with the educational process through inclusion of project themes in the curricula at the Institute of Technology, the Medical School, and the College of Education and Human Development at the University of Minnesota and the creation of a program with annual workshops, tutorials, web pages and a wiki on knowledge discovery and behavioral analysis. The team will develop an interactive exhibit for children at the Bakken Museum, and create of new instructional material for student teachers at the Institute of Child Development and similar institutions. Development of a central web repository will insure that the algorithms and the data will be readily available for appropriate research.

This project revolves around tumbling which is an exciting area of robotic locomotion that takes advantage of ground-body interactions to achieve high mobility on smaller scales when compared to conventional methods. Additionally, the required hardware complexity to produce such locomotion is very low. In this respect, tumbling can be viewed as a minimalistic approach to producing miniature mobile robots capable of traversing complex and dynamic terrains. Due to the nature of tumbling however, the added mobility comes at the price of increased control complexity. The minimalistic nature of tumbling robots generally results in underactuated systems that exhibit nonholonomic constraints which greatly complicate the motion planning problem. Additionally, tumbling often involves time-varying supports and sliding contacts with the ground. Ultimately, this research views tumbling as a largely unexplored yet promising area of research. This work addresses the intricacies of tumbling locomotion. Specifically, we are developing general planning algorithms for tumbling robots and identify important design characteristics of tumbling robots that lead to simplified control.

Seminars and workshops to bring together practitioners, end-users, researchers, and policy makers will be organized to have the maximal impact. Web-based dissemination of the algorithms and rapid prototyping/simulation tools ensure that the results of this project reach all communities. Students trained in this project participate in the US FIRST competitions, summer mentoring programs for high school students, summer schools in robotics, and other outreach programs.

Proposal #: 10-61489
PI(s): Papanikolopoulos, Nikolaos;Hondzo, Miki; Morellas, Vassilios; Anaraki, Siavash Pourmakamali; Voyles, Richard M.
Institution: University of Minnesota
Title: MRI RAPID: Collaborative Research: Swarms of Robotic Aquapods to Assess Impact of Oil Spills on Marshlands
Project Proposed:
The project, testing the adequateness of underwater robots to swiftly and repeatedly sample large areas of Golf Oil Spill, aims to determine the spatial heterogeneity of the impacts on the shoreline, revealing "hot spots." These are areas of high oil concentration that constitute a challenge since hot-spots move due to shoreline morphology, wind patterns, and water circulation. To this end, small robots called "Aquapods" will be deployed. Aquapods are miniature robots with a high mobility-to-size ratio. As a form of locomotion, they are based on tumbling which allows them to locomote on the water, under the water, and on sandy and marshy shoreline. A recent version of the Aquapod (developed as part of the IUCRC on Safety, Security, and Rescue) can be completely submerged in water to operate on a lake or stream floor. Additionally, this robot is equipped with a buoyancy control unit that allows the robot to either sink or float in water, thus offering many unique applications in both environmental monitoring and surveillance. The work develops a more advanced system than the first generation radio controlled design, incorporating functionalities more appropriate to monitor oil spill effects.
Broader Impacts:
As the proposed research can drastically improve oil spill cleanup efforts, the potential broader impacts are large. The project exhibits the ability to assess the environmental impact on areas not easily accessible by humans. The PIs have a solid plan to involve middle-schoolers from underrepresented groups in the proposed project. Moreover, the developed instrument will be free for use to the interested groups of scientists. Planned are solid dissemination and education efforts both at the University of Minnesota and the University of Denver.

This proposal seeks funding for the Center for Safety, Security, and Rescue Research studies conducted by the University of Minnesota (lead) and its research partner at the University of Denver. Funding Requests for Fundamental Research are authorized by an NSF approved solicitation, NSF 10-507. The solicitation invites I/UCRCs to submit proposals for support of industry-defined fundamental research.

The proposal is about further developing ongoing robot design activities for safety and security. The presented research focus is in the area of underwater robot design, and it describes activities regarding the waterproofing, buoyancy control, camera stabilization and sensor fusion strategies. The proposed work builds on the strong previous track record, and the team is strong. Unlike most existing approaches, the proposed effort should provide miniature size robots with high mobility.

This proposal presents a set of robotic systems (including an amphibious, tumbling robot) and algorithms that capitalize on earlier pertinent work at the Industry/University Cooperative Research Center for Safety, Security, and Rescue Research. The proposal also offers innovative stabilization algorithms and hardware for the sensor suite in order to improve the quality of data provided to the first responder. In addition, innovative sensor fusion schemes for underwater navigation are proposed. The broader impacts of the proposed work range from a planned inclusion of undergraduates into the research plan to the possibility that life-saving commercial devices could eventually result from the work described. The connection of this work with the NSF Safety, Security, and Rescue Research Center will ensure that the proposed technologies will be exposed to first responders at the proposed field trials.

Safety, Security and Rescue Research Center
Proposal #1127947
Proposal #1127938

This proposal seeks funding for the Safety, Security and Rescue Research Center sites at the University of Denver and the University of Minnesota. Funding Requests for Fundamental Research are authorized by an NSF approved solicitation, NSF 10-601. The solicitation invites I/UCRCs to submit proposals for support of industry-defined fundamental research.

Use of robots teams in extreme environments for wilderness search and rescue, urban search and rescue and surveillance are anticipated to require operation on land, water and under water. The proposed effort will explore means to maintain coordination of such multiple robot teams through environmental disturbances, communication lapses, and hardware and software failures. Moreover the work to establish resilient robot teams will enable the exploration of emergent behavior of the team that has the potential to arise due to these various faults. The effort will establish a real-time test bed based on the Aquapod robot and include development of test methodologies for single and multiple robots in various aqueous environments.

The outcomes of the proposed work have the potential for significant impact in the areas of public safety and search and rescue across civilian and military application spaces. The work is supported by the Industry Advisory Board as well as individual industry members of the center and has the potential to extend the center?s portfolio through expansion into the areas of robot teams, their behavior and fault tolerant control. The impact of the work has the potential to extend beyond the center by virtue of the testing methodology to be developed for robot teams. The center will involve graduate students and undergraduates in the work and incorporate the results in existing coursework in robotics and software systems engineering.

Proposal #: CNS 11-38020
PI(s): Papanikolopoulos, Nikolaos
Institution: University of Minnesota
Title: RAPID: Robotic Systems for Disaster Assessment and Disaster Relief
Project Proposed:
This RAPID project proposal, consisting of a development effort and dispatching of improved robotic inspection tools to colleagues in Japan, will expand on existing, successful robotic development efforts to create a new robot with specific capabilities needed to assist response to the Sendai earthquake and tsunami. The novelty of the proposed robot may be found in its locomotion capabilities and its resistance to dirt and water. Its ease of use make it stand out among research artifacts. The researchers will be engaged in some of the following activities:
- Engineer a treaded version of the Aquapod submersible robot
- Develop a sensor suite for the Aquapod
- Develop a simulator tool for first responders
- Deploy the robots in Japan through Japanese colleagues
The investigators collaborate with Dr. Kazuhiro Kosuge from Tohuku University and Satoshi Takoro from Tohohu University. Their letters and other appropriate invitations have been secured from the Japanese researchers and responders, including the International Rescue Systems Institute, to ship a small team of robots to Sendai, Japan to assist in the inspection of critical infrastructure and other recovery operations as well as perform research on the efficacy of emergency response methods and practices. (The search and rescue phase has been terminated.) These new, improved robots combine treads and limbs for novel locomotion capabilities for their sensor payloads. New sensor payloads enable expanded sensing of location, environmental data, and radiation. With these enhanced capabilities, a team of robots will be deployed for victim and economic recovery decision-making while simultaneously collecting ephemeral data not possible with existing devices. The Sendai disaster is unique in its large geographical and economic scale and types of damage. These robots deployed in Sendai will provide an additional modality of unmanned systems to be tested along with conventional robotic types.
Broader Impacts:
This proposal promises an immediate benefit to society by supporting economic recovery efforts in Japan through a participatory research paradigm. Moreover, long term benefits for future disasters are in evidence since emergency response and unmanned systems are both formative domains and the data collected will advance the discovery and understanding of intelligent, human-centered systems in unpredictable situations. The PI has a history of outreach to museum, K-12, and general public education events. In addition, the simulator tool promises to enhance the safety of emergency responders, which has multiplicative benefits.

This Innovation Corps project proposes to test the commercial feasibility of a robot that can move on the ground, fly, and hover. In the ground mode, the rotors and stabilizer bar are folded down along the length of the robot's body. It transforms into its flight mode by positioning itself on-end, with its long axis oriented vertically rather than horizontally, and unfolding its flight mechanisms. This type of robot offers several key benefits: 1) Capability to carry out complex missions, and return to home base, 2) Un-assisted take off (ability to switch locomotion modes at will), 3) Hovering (maintaining position in the air), 4) Ability to scale large obstacles and fly over rough terrain, and 5) Efficient ground-mode travel. While some of these benefits could be realized with a small helicopter, the fourth item, "efficient ground-mode travel", sets this design apart by giving it the ability to conserve energy while still making progress toward its objective.

A robot that has the capability to switch air and ground modes at will as many times as necessary to complete a mission, with "stop-and-go" capabilities both on the ground in the air, enables adaption to diverse missions, uncontrollable environments, and unpredictable threats. Many missions require movement through terrain which is not easily traversable, such as in an earthquake search-and-rescue scenario. In addition to search-and-rescue applications, this type of robot would be ideal for many other applications such as: covert surveillance, perimeter surveillance, reconnaissance, law enforcement, hostage scenarios, special force operations, hazardous and/or contaminated (nuclear, biological, chemical, or radiological) area inspection, and any mission which requires situational awareness over the line-of-sight or over the line-of-reach.

The proposed research proposes to develop novel image processing and machine learning algorithms for the detection and segmentation of cancerous regions from high-resolution images of tissue slides, distinguishing them from healthy/benign regions. The proposed research plans to advance (i) A reliable framework for use in the accurate segmentation of cancerous regions in tissue slides; (ii) New algorithms for texture analysis; (iii) Innovative representations that can increase the system throughput; (iv) Effective machine learning techniques and transparent user interfaces to assist in the reduction of the time that a pathologist needs to examine each slide; and (v) Extensive testing to a variety of cancers including prostate and breast cancer.

The proposed work has the potential to yield algorithms to facilitate the design of systems that can increase the likelihood of cancer detection. The work is supported by the Industry Advisory Board as well as individual industry members of the center and has the potential to extend the center?s portfolio. The PIs plan to introduce the research content into summer robotic camps for middle schoolers from underrepresented groups, create a website along with a wiki, visit local groups and K-12 schools, and include the research products into the curricula of the participating institutions.

Proposal #: 13-38042
PI(s): Papanikolopoulos, Nikolaos
Banerjee, Arindam; Bernstein, Gail; Hadjiyanni, Tassoulla; Lim, Kelvin, K.
Institution: University of Minnesota-Twin Cities
Title: MRI/Dev.: Instrument that Monitors Behaviors Associated with OCD and Schizophrenia
Project Proposed:
This project, developing a new instrument to facilitate data collection associated with clinical assessment of complex mental health disorders such as obsessive-compulsive behaviors (known as OCD), aims to enable long-term research advances in computer vision, activity recognition and tridimensional reconstruction algorithms to automate the identification of behaviors typical of OCD subjects. The immersion of selected subjects in a virtual reality room (CAVE) is used to trigger specific behaviors to be captured and analyzed. The sophisticated sensor system under development will serve to collect these data and provide intelligent data processing capabilities that would enable future exploration and testing of new diagnostic and therapeutic protocols, leading to the establishment of the basis for, and utility in, seeking early at-risk markers in children and adolescents. This instrument initiative is based on the premise that expertise can accurately identify useful diagnostic markers and on the belief that technologies can now be developed to collect massive behavioral data in ways not previously done and discover behavioral patterns.
The instrument is expected to
- Provide extensive data collection associated with subjects diagnosed with the respective disorders (data useful not only to clinicians but also to computer vision and machine learning researchers among others),
- Capture interactions, behaviors, and physiological reactions to real and/or synthetic multimodal stimuli (optical, acoustical, etc.),
- Allow computer and information scientists to develop computational tools and algorithms to generate quantitative, adequate, and cost-effective norms for screening a broad population,
- Enhance Cognitive Behavior Therapy (CBT) procedures and diagnostic protocols by integration of technologies that can excite or inhibit triggers for schizophrenic or OCD episodes,
- Assess particular Augmented and Virtual Environments (AE/VEs) and social media devices (smart phones and tablets) and their impacts on the cognitive presence of normal versus afflicted subjects, and
- Evaluate whether an enhanced cognitive presence via an AE/VE can increase or suppress (habituate) the intensity of behavioral symptoms detectable by sensors.
Broader Impacts: Among these we have:
- Creation of large and complex datasets that will enable computer scientists to apply the newest computational tools on them,
Development of a potentially transformative technology-driven instrument for detecting early risk markers of OCD,
- Exploration of a platform well-suited to new directions for a better characterization of mental disorders,
- Systematic database development of quantified, multimodal data and a sounder and more precise basis for earlier detection,
- Reduction of overall costs and a parallelizing reduction in the long-term costs due to previously delayed or incorrect diagnoses,
- Reduction of anxiety, disruption, stress, and sometimes real tragedy on patients and their families,
- Earlier detection and reduced need for drug-based, later stage interventions enabled by the ease of testing, and associated societal benefits, and
- Student education and training in the use of the instrument.

The proposed work seeks to enable plug and play use of multiple platforms in the development of multi-robot teams that interact with human teams. The research will develop a software engineering framework viewed as crucial for the proliferation of intelligent robots. The proposed approach will (i) Develop of a framework that allows quick integration of new robot designs into the robotic team; (ii) Develop a user-friendly human-team interface, (iii) Create methods to perform asset (robot) management, (iv) Software test to check the viability of the proposed solutions, and (v) Experimentally validate the system in real-world settings in consultation with center members.

Robot teams have the potential for significant societal impact given their design and implementation can be made more efficient and low cost. Interchangeable use of hardware and software to achieve a wide range of functions is critical to achieving this potential and the associated economic impact that will accompany it. The proposed software framework for robotics is supported by the Industry Advisory Board and has the potential to extend the center?s portfolio significantly through expansion into the area of robotic team system software engineering. The research will impact and be used in the center?s annual robotics camp and other outreach activities.

The project develops and tests novel compressive sensing and sensor locating techniques that are adaptable to a myriad of different mobile robot designs while operable on today's wireless communication infrastructures. Unique in-situ laboratory and field experiments provide tangible results to scientists and other stakeholders that can be leveraged to advance these systems into future real-world hazard management scenarios. The research team develops new technological approaches that results in mobilizing more intelligent, automated "eyes and ears on the ground." Outreach efforts include: (i) integration of the activities with practitioners; (ii) Seminars/webcasts to audiences like environmental engineers and first responders; (iii) Annual technology day camps to attract middle-schoolers from under-represented groups to engineering; (iv) Demonstrations to local K-12 institutions; (v) Inclusion of the project themes to the regular curricula; and (vi) International collaborations.

This project introduces Robotics 2.0; a framework that targets autonomous robots that are co-workers and co-protectors, adapting to and working with humans. The research team develops a Cyber-Control Network (CCN) to allow multiple fixed and mobile robotic environmental sensing and measurements to adapt quickly to the changing environment by dynamically linking sub-networks of actuation, sensing, and control together. The design of such CCN ControlWare, and compressive sensing architectures, could be adapted to other large-scale problems beyond disaster response, mitigation, and management, such as power grid monitoring and reconfiguration, or regional urban traffic operations to respond to traffic congestion and incidents. The robotic sensing platforms do not require a-priori knowledge of the hazardous and dynamically changing environments they are monitoring. The Robotics 2.0 framework allows to swiftly respond, to prepare, and to manage various types of disasters.

This project wants to capitalize on a team of robots that move in order to better sense the environment and then perform basic manipulation tasks. The vision of the project is to integrate robots easily, provide human-team interfaces, and develop manipulation algorithms. The research will involve the development of perception strategies and manipulation schemes that will allow operation of the robot teams in real-world environments during search and rescue missions. In addition, this research will involve working in cluttered scenes where the lighting conditions may not be ideal. The project addresses: (i) Research on the novel problem of robotic perception and manipulation of target objects that interact with other objects as an integral part of the environment, which cannot be fully isolated in views and in physical arrangements before being manipulated; (ii) An appearance-based approach for recognition and pose estimation of 3D objects in cluttered scenes from a single view; (iii) Development of a measure of scene recognizability from each viewpoint to evaluate how accurately partially-occluded objects are recognized and how well their poses are estimated; (iv) Creation of solutions for disassembly analysis of 3D structures, extending our preliminary analysis of 2D structures; (v) Development of grasping based on the results of perception and with the aid of stability analysis of the arrangement of the objects and their interaction with the environment and with one another; and (vi) Experimental validation of the system in real-world settings, in close consultation with our industrial partners.

This project will allow the creation of manipulation capabilities along with perception schemes to facilitate the development of a multi-robot team for search and rescue missions. The impacts of the project include: (i) Expansion of the annual robot summer camp to include robot-team activities with the objective of attracting middle-schoolers from under-represented groups to computer science/electrical engineering; (ii) Integration of the activities with first responders through the UPenn and UNC Charlotte collaborations; (ii) Experimental validation in SAFL at UMN and the Disaster City in TX; (iii) Offering project themes to REU undergraduates or to the UROP program; (iv) Outreach programs that involve demonstrations to local K-12 institutions; and (v) Inclusion of the project theme to the regular curricula.

In the recent past the term "Smart Cities" was introduced to mainly characterize the integration into our daily lives of the latest advancements in technology and information. Although there is no standardized definition of Smart Cities, what is certain is that it touches upon many different domains that affect a city's physical and social capital. Smart cities are intertwined with traffic control systems that use advanced infrastructures to mitigate congestion and improve safety. Traffic control management strategies have been largely focused on improving vehicular traffic flows on highways and freeways but arterials have not been used properly and pedestrians are mostly ignored. This work proposes to introduce a novel hierarchical adaptive controls paradigm to urban network traffic control that will adapt to changing movement and interaction behaviors from multiple entities (vehicles, public transport modes, bicyclists, and pedestrians). Such a paradigm will leverage several key ideas of cyber-physical systems to rapidly and automatically pin-point and respond to urban arterial congestion thereby improving travel time and reliability for all modes. Safety will also be improved since advanced warnings actuated by the proposed cyber-physical system will alert drivers to congested areas thereby allowing them to avoid these areas, or to adapt their driving habits. Such findings have a tangible effect on the well-being, productivity, and health of the traveling public.

The primary goal is to create a Cyber-Control Network (CCN) that will integrate seamlessly across heterogeneous sensory data in order to create effective control schemes and actuation sequences. Accordingly, this project introduces a Cyber-Physical architecture that will then integrate: (i) a sub-network of heterogeneous sensors, (ii) a decision control substrate, and (iii) a sub-actuation network that carries out the decisions of the control substrate (traffic control signals, changeable message signs). This is a major departure from more prevalent centralized Supervisory Control And Data Acquisition (SCADA), in that the CCN will use a hierarchical architecture that will dynamically instantiate the sub-networks together to respond rapidly to changing cyber-physical interactions. Such an approach allows the cyber-physical system to adapt in real-time to salient traffic events occurring at different scales of time and space. The work will consequently introduce a ControlWare module to realize such dynamic sub-network reconfiguration and provide decision signal outputs to the actuation network. A secondary, complementary goal is to develop a heterogeneous sensor network to reliably and accurately monitor and identify salient arterial traffic events. Other impacts of the project include the integration of the activities with practitioners (e.g., traffic engineers), annual workshops/tutorials, and outreach to K-12 institutions.

Part 1:
This project is about the organization of an international symposium that deals with computational tools for mental health assessment. Early intervention can dramatically improve the individual's quality of life, for many psychiatric disorders. Accounting for 25% of all years of life lost to disability and premature mortality, mental health disorders are the leading cause of disability in the United States and Canada, according to the World Health Organization. Symptoms of mental illness which emerge in childhood and early adolescence are actually the later stages of a process which began years earlier. Hence, psychiatric research is immensely interested in identifying and detecting risk-markers (genetic, neural, behavioral, and/or social deviations) that indicate elevated risk for the development of specific mental illnesses prior to the onset of symptoms. This meeting will be organized at Cyprus where mental health issues are still not addressed appropriately due to cultural issues. The meeting will bring US experts and graduate students together with researchers from University of Cyprus, where new efforts are initiated to deal with the problem by scientists involved with psychiatry and computer science. The complementary skills of CS, ECE, and medical experts from the two countries will benefit this novel research area, and ultimately benefit the U.S and the world. The discussions will be focused on tackling the gap in the current state of the art and developing strategies to minimize that gap. This includes identifying the progress that has already been made while recognizing the pitfalls and determining new opportunities for mental health monitoring using vision and sensor network technologies.

Part 2:
This project is about the organization of an international meeting that will involve the creation of computational algorithms and intelligent distributed systems to assist with the early diagnosis of individuals who are at risk of developing behavioral disorders. Previous research has indicated that two critical areas of behavioral investigation for use in identifying at-risk individuals have been abnormalities in motor activities and emotional range displays, especially of the face. Motor abnormalities are based on the observation that motor control involves the circuits of the brain associated with dopamine, which are also implicated in behavioral disorders. Many different disorders share the observation of disruption in the emotional range regulation, so monitoring of facial expressions should be discussed.

To date, assessments of motor and emotional range have been done by the experts who view and rate videos of an individual. However, these experts? subjective ratings limit the analysis of behavioral conditions to only a narrow range of behaviors, work only for small populations of individual subjects, and are both costly and dependent on the observer's particular expertise. In order to enable a wider population screening, automation is required. The proposed meeting is focused on the development of this automated process that uses vision and sensor networks. Innovative ways of capturing and quantifying the expertise of experts will be presented along with metrics for assessing the evolution of the behavior. In addition, new computational tools that support the evaluation of the effectiveness of interventions will be discussed.

This project, developing a high efficiency solar power enabled aerial instrument that incorporates sensory and processing requirements into the design methodology, focuses on the development of a (robotic) instrument that fills a niche in certain applications (i.e., small scale solar powered Unmanned Aerial Vehicles (UAVs)). The work involves the creation of a platform for real-world information gathering applications with special focus on robust navigation, distributed sensing, and collaborative scenarios. In particular, the development of a small scale solar powered UAV provides a robotic platform capable of communication, sensory coverage, and flight endurance unattainable through traditional UAV design. Existing UAVs utilizing solar power are constrained to large aircraft designs requiring a traditional runway for takeoff and landing. In contrast, electric powered small scale UAVs suffer from minimal flight time. Additionally, aerial robots provide significant advantages over ground based systems as they are unaffected by terrain and obstacles, and provide an information-rich vantage point from the air. Conventional aerial robots are significantly limited by their flight time and therefore their deployment is severely restricted. Flight time has been the central limitation for airborne sensory information and has prevented experimental research and real-world applications from being performed. The development of a high efficiency aircraft that leverages solar energy as a resource provides a solution to the flight time problem.

The methodologies mentioned involve the incorporation of hierarchical planning methods that include high-level reasoning for the optimal number of UAVs/sensors required, and low-level reasoning for efficient sensor placement throughout the environment. Specifically, the use of solar powered UAV platforms will enable work in the areas of precision agriculture and environmental science, requiring strong demand for high endurance systems capable of supporting a variety of sensor and measurement units. (For example, timely and repetitive relevant information regarding crop health is necessary for corrective actions to be implemented.) Additionally, collaborative operation of dynamically placed sensor platforms and devices can only be enabled through the use of small solar powered airplanes like the ones to be implemented. The instrument that this project funds enables work in application areas that require continuous and repetitive operation such as Energy, Environment, Agriculture, etc. Consequently, this work addresses the
- Development of a small scale solar powered platform that is capable of multi-day flight,
- Experimental validation of the scaled down solar UAV instrument,
- Long-term solar powered flight planning based on sensory data collected, and
- Creation of benchmarks that will allow comprehensive evaluation of the small solar powered UAV instrument.

Proposal #: 15-14626
PI(s): Papanikolopoulos, Nikolaos; Lim, Kelvin O.
Institution: University of Minnesota-Twin Cities (UMN)
Title: RAPID: Remote Monitoring of Ebola Patients and Their Treating Physicians
Project Proposed:
This project, melding proven technologies, aims to assist in an immediate challenge now facing those engaged in direct patient care in the ongoing Ebola epidemic. Namely, employing interactive machine vision technologies and intelligently encoded protective gear to proactively assist health care providers (e.g., adhering to their required NIH/CDC (Center for Disease Control) Safety Procedures-the explicit movement sequences required for the proper donning of sterilized, protective coverings, and afterwards for the safe, step-wise and controlled removal and subsequent disposition of these now potentially lethal items). The crucial importance of this very high standard was underscored recently by the infection of two Dallas-area nurses where protocols were thought breached, and then by the CDC?s response, which next imposed an enhanced buddy system to better ensure future compliance. Notwithstanding, human monitors can sometimes lapse, especially when procedures become routine, as time pressures mount. Fortunately, incorporating respective lessons learned in the creation of an automated and interactive buddy is a readily achievable next step, given the advances in automated human activity and gesture recognition, with experience of the strict procedures evolved in conjunction with tele-operational assistance methods for domains such as surgery, and with modern graphical and human-friendly man-machine interfaces. Obviously, this new knowledge serves a useful purpose.
Broader Impacts:
The relatively low cost in the longer term should be noted. Even portable extensions to any environment can be disproportionally costly or even catastrophic, particularly when participants fail to adhere to strict standards (usually tedious and unwelcome). In industry (e.g., clean rooms, nuclear sites, etc.), in medicine (e.g., operating theaters), in the military, and in some research areas, the cost of recording and registering sessions that compel the full compliance to required safety measures might easily be offset by better outcomes, fewer accidents and reduced insurance rates. This project contributes in building a coalition of large NSF awardees to jointly address public health problems of national and global significance using the state of the art in Computer Science. The UMN team will assist in the urgent data collection regarding the effectiveness of the protective gear used in Ebola-infected regions.

This research will enable the development of new quantitative crisis maps to approximate local regional rates of infectious Ebola virus spreading in near real time thereby enhancing the effective distribution of limited availability of recently developed vaccines to mitigate the spread of Ebola. Furthermore, it will provide a mechanism for validating and improving operational forecast models of the CDC by comparisons with regional social media observations. It may allow for disease surveillance and early detection among our third world countries. Moreover, this project is expected to be generalizable to other infectious virus conditions; the research enables exploiting social media health data from future hand held devices. Although enough speculation exists about the role of the gear, little data is available to push for changes in the protocols or the gear itself. Beyond the present issue of Ebola, the proposed methodology and the coalition building effort will support solutions in a wide range of public health issues.

This project, developing an instrument to enable a pioneering way to canonically represent and systematically quantify behavioral manifestations for a suite of neuropsychiatric disorders, integrates a powerful computer cluster with various sensor modalities and associated immersive technology components that facilitate data analysis and perception of physical and virtual experiences. The instrument will bring together neuropsychiatrists and engineering scientists seeking the development and deployment of tools that promote discovery and new knowledge in their respective domains, and innovation with immediate societal impact. It aims to enable medical experts to explore clinical hypotheses during diagnosis and treatment. Concurrently, scientists/engineers will explore the implementation of new computational tools and algorithms that satisfy new requirements utilizing a set of clinical hypotheses.

Challenging the current understanding of a spectrum of neuropsychiatric conditions and triggering the explorations of new ways to affect them, this instrument should contribute to our understanding of the presence and evolution of mental states. Its unique characteristics will help in integrating a range of informational cues (e.g., visual, acoustic, haptic) along with virtual simulated information. This interplay between physical and virtual experiences and stimuli and their perception in an immersive environment constitutes the ultimate objective which will unleash creativity in better understanding and treat mental illnesses.

The tight coupling of hardware and software components will enhance the various types of interactions such as expert-patient, and patient-virtual objects. Specifically, the instrument will support the following research efforts:
- Comprehensive compilation of behavioral data,
- Development of visualization tools,
- Delivery of haptic information feedback,
- Creation and interaction of virtual objects, and
- Enhanced sentiment understanding from audio/visual signals, facial expression representation and analysis.
The instrument enables the creation of communication channels with clinics across the country. Moreover, it contributes in forming a strong foundation to explore and understand a broad variety of neuropsychiatric disorders manifested via verbal and non-verbal communication pathways, aims to form a foundation that promotes strong collaboration targeting real-world challenges with immediate societal and economic impact, and offers an environment for discovery via the strong coupling of data visualization, User Interfaces-User eXperience (UI-UX) design, immersive perception, data exploration, and machine learning.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

This project represents a step toward a computational model capable of detecting early social behavioral markers in children at risk for autism spectrum disorder, schizophrenia, and obsessive-compulsive disorder. The real-time 3D Social Signal Imaging System (SSIS) will be designed to precisely measure social signals utilizing cameras producing billions of pixels dozens of times per second. The infrastructure will be designed to enable reconstruction of the 3D geometry of gaze, face, finger, body, and physical appearance. The system is expected to be capable of generating a vast amount of multiple perspective visual data to reconstruct high fidelity 3D signals, needed to enable social intelligence that can decode every nuance of human expression.

The ability to discern subtle social signals (e.g., gaze following) can be computationally modeled by leveraging a massive camera system. The Social Signal Imaging System (SSIS) facilitates quantitative measurements of the social signals in 3D at unprecedented temporal and spatial resolutions. This development involves the following steps: (i) Design a distributed visual computing architecture to efficiently process the Multiview visual data streams; (ii) Build a new high-fidelity 3D representation of the view-invariant social signals (gaze, face, finger, body, appearance); (iii) Create a novel 3D dataset of social signals for use in discovering behavioral markers; and (iv) Develop new computer vision algorithms (recognition, matching, tracking, reconstruction) tailored to social signal imaging that minimize computational latency while maintaining accuracy. The system provides a unique characterization of microscopic social signals that enable overcoming fundamental limitations of existing approaches in behavioral assessment of at-risk children. This work impacts diverse disciplines such as robotics, neuroscience, psychology, psychiatry, and medicine. The outcomes will be disseminated through K-12 students from under-represented groups via workshops, machine learning and technology summer camps, and other activities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.