The objective of this research is to apply grammatical inference models recently developed in the field of linguistics and phonology, as a basis for abstraction, composition, symbolic control, and learning in distributed multi-agent cyber-physical systems. The approach is to map the system dynamics, specifications, and task interdependences to finite abstract models, and then describe the desired behavior of the system in an appropriate grammar that can be decomposed into local agent specifications. In this framework, the agents can learn the behavior of their environment by observing its dynamics, and update their specifications accordingly. The proposed approach to learning in cyber-physical systems, which is based on grammatical inference at a purely discrete level, is a significant departure from current works. Following this approach, one can reason about large-scale processes resulting from event interdependencies between agents, without having to construct large product systems. To realize this plan, specific technical advances on modeling, abstraction, and control synthesis are proposed. Questions related to formally factoring and composing heterogeneous systems are pervasive in the fields of formal languages and computational learning. There are also applications of commercial significance in the area of discovering new azeotropic mixtures based on documented pairs of compounds that are known to have the particular property. Proposed dissemination and outreach activities include the involvement of middle and high school students and teachers, integrated in existing NSF-sponsored programs at the University of Delaware and Boston University.

The objective of this research is to apply grammatical inference models recently developed in the field of linguistics and phonology, as a basis for abstraction, composition, symbolic control, and learning in distributed multi-agent cyber-physical systems. The approach is to map the system dynamics, specifications, and task interdependences to finite abstract models, and then describe the desired behavior of the system in an appropriate grammar that can be decomposed into local agent specifications. In this framework, the agents can learn the behavior of their environment by observing its dynamics, and update their specifications accordingly. The proposed approach to learning in cyber-physical systems, which is based on grammatical inference at a purely discrete level, is a significant departure from current works. Following this approach, one can reason about large-scale processes resulting from event interdependencies between agents, without having to construct large product systems. To realize this plan, specific technical advances on modeling, abstraction, and control synthesis are proposed. Questions related to formally factoring and composing heterogeneous systems are pervasive in the fields of formal languages and computational learning. There are also applications of commercial significance in the area of discovering new azeotropic mixtures based on documented pairs of compounds that are known to have the particular property. Proposed dissemination and outreach activities include the involvement of middle and high school students and teachers, integrated in existing NSF-sponsored programs at the University of Delaware and Boston University.

The objective of this research is to develop an atomic force microscope based cyber-physical system that can enable automated, robust and efficient assembly of nanoscale components such as nanoparticles, carbon nanotubes, nanowires and DNAs into nanodevices. The approach in this project is based on the premise that automated, robust and efficient nanoassembly can be achieved through tip based pushing in an atomic force microscope with intermittent local scanning of nanoscale components. In particular, in order to resolve temporally and spatially continuous movement of nanoscale components under tip pushing, the research is exploring the combination of intermittent local scanning and interval non-uniform rational B-spline based isogeometric analysis in this research. Successful completion of this research is expected to lead to foundational theories and algorithmic infrastructures for effective integration of physical operations (pushing and scanning) and computation (planning and simulation) for robust, efficient and automated nanoassembly. The resulting theories and algorithms will also be applicable to a broader set of cyber physical systems. If successful, this research will lead to leap progress in nanoscale assembly, from prototype demonstration to large-scale manufacturing. Through its integrated research, education and outreach activities, this project is providing experiences and understanding in cyber-physical systems and nanoassembly for students from high schools to graduate schools. The goal is to increase interest in science and engineering among domestic students and therefore strengthen our competitiveness in the global workforce.

The objective of this research is to develop advanced distributed monitoring and control systems for civil infrastructure. The approach uses a cyber-physical co-design of wireless sensor-actuator networks and structural monitoring and control algorithms. The unified cyber-physical system architecture and abstractions employ reusable middleware services to develop hierarchical structural monitoring and control systems. The intellectual merit of this multi-disciplinary research includes (1) a unified middleware architecture and abstractions for hierarchical sensing and control; (2) a reusable middleware service library for hierarchical structural monitoring and control; (3) customizable time synchronization and synchronized sensing routines; (4) a holistic energy management scheme that maps structural monitoring and control onto a distributed wireless sensor-actuator architecture; (5) dynamic sensor and actuator activation strategies to optimize for the requirements of monitoring, computing, and control; and (6) deployment and empirical validation of structural health monitoring and control systems on representative lab structures and in-service multi-span bridges. While the system constitutes a case study, it will enable the development of general principles that would be applicable to a broad range of engineering cyber-physical systems. This research will result in a reduction in the lifecycle costs and risks related to our civil infrastructure. The multi-disciplinary team will disseminate results throughout the international research community through open-source software and sensor board hardware. Education and outreach activities will be held in conjunction with the Asia-Pacific Summer School in Smart Structures Technology jointly hosted by the US, Japan, China, and Korea.

The objective of this research is the development of methods and software that will allow robots to detect and localize objects using Active Vision and develop descriptions of their visual appearance in terms of shape primitives. The approach is bio inspired and consists of three novel components. First, the robot will actively search the space of interest using an attention mechanism consisting of filters tuned to the appearance of objects. Second, an anthropomorphic segmentation mechanism will be used. The robot will fixate at a point within the attended area and segment the surface containing the fixation point, using contours and depth information from motion and stereo. Finally, a description of the segmented object, in terms of the contours of its visible surfaces and a qualitative description of their 3D shape will be developed. The intellectual merit of the proposed approach comes from the bio-inspired design and the interaction of visual learning with advanced behavior. The availability of filters will allow the triggering of contextual models that work in a top-down fashion meeting at some point the bottom-up low-level processes. Thus, the approach defines, for the first time, the meeting point where perception happens. The broader impacts of the proposed effort stem from the general usability of the proposed components. Adding top-down attention and segmentation capabilities to robots that can navigate and manipulate, will enable many technologies, for example household robots or assistive robots for the care of the elders, or robots in manufacturing, space exploration and education.

The objective of the research is to develop tools for comprehensive design and optimization of air traffic flow management capabilities at multiple spatial and temporal resolutions: a national airspace-wide scale and one-day time horizon (strategic time-frame); and at a regional scale (of one or a few Centers) and a two-hour time horizon (tactical time-frame). The approach is to develop a suite of tools for designing complex multi-scale dynamical networks, and in turn to use these tools to comprehensively address the strategic-to-tactical traffic flow management problem. The two directions in tool development include 1) the meshed modeling/design of flow- and queueing-networks under network topology variation for cyber- and physical- resource allocation, and 2) large-scale network simulation and numerical analysis. This research will yield aggregate modeling, management design, and validation tools for multi-scale dynamical infrastructure networks, and comprehensive solutions for national-wide strategic-to-tactical traffic flow management using these tools. The broader impact of the research lies in the significant improvement in cost and equity that may be achieved by the National Airspace System customers, and in the introduction of systematic tools for infrastructure-network design that will have impact not only in transportation but in fields such as electric power network control and health-infrastructure design. The development of an Infrastructure Network Ideas Cluster will enhance inter-disciplinary collaboration on the project topics and discussion of their potential societal impact. Activities of the cluster include cross-university undergraduate research training, seminars on technological and societal-impact aspects of the project, and new course development.

The objective of this research is the creation of a coastal observing system that enables dense, in situ, 4D sensing through networked, sensor-equipped underwater drifters. The approach is to develop the technologies required to deploy a swarm of autonomous buoyancy controlled drifters, which are vehicles that can control their depth, but are otherwise carried entirely by the ocean currents. Such Lagrangian sampling promises to deliver a wealth of new data, ranging from applications in physical oceanography (mapping 3D currents), biology (observing the dispersion of larvae and nutrients), environmental science (tracking coastal pollutants and effluents from storm drains), and security (monitoring harbors and ports). This observing system fundamentally requires accurate positions of the drifters (to interpret the spatial correlations of data samples), swarm control algorithms (to achieve desired sampling topologies), and wireless communication (to coordinate between the individual drifters). This research will create distributed techniques to self-localize the drifter swarm, novel swarm control algorithms that enable topology manipulation while purely leveraging the stratified flow environment, and efficient wireless underwater communication for information sharing. This project has significant societal impact and educational elements. Underwater drifter swarms will enable novel insights into a wide array of scientific questions, including understanding plankton transport, accumulation and dispersion as well as monitoring harmful algal blooms. Undergraduates will play an active role in many aspects of this project, thereby offering them a uniquely interdisciplinary experience. Finally, outreach to high school students will occur through the UCSD COSMOS summer program.

The objective of this research is to develop algorithms and software for treatment planning in intensity modulated radiation therapy under assumption of tumor and healthy organs motion. The current approach to addressing tumor motion in radiation therapy is to treat it as a problem and not as a therapeutic opportunity. However, it is possible that during tumor and healthy organs motion the tumor is better exposed for treatment, allowing for the prescribed dose treatment of the tumor (target) while reducing the exposure of healthy organs to radiation. The approach is to treat tumor and healthy organs motion as an opportunity to improve the treatment outcome, rather than as an obstacle that needs to be overcome. Intellectual Merit: The leading intellectual merit of this proposal is to develop treatment planning and delivery algorithms for motion-optimized intensity modulated radiation therapy that exploit differential organ motion to provide a dose distribution that surpasses the static case. This work will show that the proposed motion-optimized IMRT treatment planning paradigm provides superior dose distributions when compared to current state-of-the art motion management protocols. Broader Impact: Successful completion of the project will mark a major step for clinical applications of intensity modulated radiation therapy and will help to improve the quality of life of many cancer patients. The results could be integrated within existing devices and could be used for training of students and practitioners. The visualization software for dose accumulation could be used to train medical students in radiation therapy treatment planning.

The objective of the research is to develop tools for comprehensive design and optimization of air traffic flow management capabilities at multiple spatial and temporal resolutions: a national airspace-wide scale and one-day time horizon (strategic time-frame); and at a regional scale (of one or a few Centers) and a two-hour time horizon (tactical time-frame). The approach is to develop a suite of tools for designing complex multi-scale dynamical networks, and in turn to use these tools to comprehensively address the strategic-to-tactical traffic flow management problem. The two directions in tool development include 1) the meshed modeling/design of flow- and queueing-networks under network topology variation for cyber- and physical- resource allocation, and 2) large-scale network simulation and numerical analysis. This research will yield aggregate modeling, management design, and validation tools for multi-scale dynamical infrastructure networks, and comprehensive solutions for national-wide strategic-to-tactical traffic flow management using these tools. The broader impact of the research lies in the significant improvement in cost and equity that may be achieved by the National Airspace System customers, and in the introduction of systematic tools for infrastructure-network design that will have impact not only in transportation but in fields such as electric power network control and health-infrastructure design. The development of an Infrastructure Network Ideas Cluster will enhance inter-disciplinary collaboration on the project topics and discussion of their potential societal impact. Activities of the cluster include cross-university undergraduate research training, seminars on technological and societal-impact aspects of the project, and new course development.

Using the newly introduced idea of a sensor lattice, this project conducts a systematic study of the "granularity'' at which the world can be sensed and how that affects the ability to accomplish common tasks with cyber-physical systems (CPSs). A sensor is viewed as a device that partitions the physical world states into measurement-invariant equivalence classes, and the sensor lattice indicates how all sensors are related. Several distinctive characteristics of the pursued approach are: 1) Virtual sensor models are developed, which correspond to minimal information requirements of common tasks and are independent of particular physical sensor implementations. 2) Uncertainty is decoupled into disturbances and pre-images, the latter of which yields the measurement-invariant equivalence classes and sensor lattice. 3) The development of particular spatial and temporal filters that are based on minimal information requirements of a task. 4) Formally establishing the conditions that enable sensors in a CPS to be interchanged, and then determining the relative complexity tradeoffs. The intellectual merit is to understand how mappings from the physical world to sensor outputs affect the solvability and complexity of commonly occurring tasks. This is a critical step in the development of mathematical and computational CPS foundations. Broader impact is expected by improving design methodologies for CPS solutions to societal problems such as assisted living, environmental monitoring, and automated agriculture. The sensor lattice approach is transformative because it represents a new paradigm with which to address basic sensor-based inference issues, which extend well beyond the traditional academic boundaries.