JVJohn Vian

John Vian

*This is an algorithmically generated author record based on publicly available information.

John Vian has a total of 35 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Scalable Accelerated Decentralized Multi-Robot Policy Search in Continuous Observation Spaces

    by Shayegan Omidshafiei, Christopher Amato, Miao Liu, Michael Everett, Jonathan P. How, John Vian

    This paper presents the first ever approach for solving \emph{continuous-observation} Decentralized Partially Observable Markov Decision Processes (Dec-POMDPs) and their semi-Markovian counterparts, Dec-POSMDPs. This contribution is especially important in robotics, where a vast number of sensors provide continuous observation data. A continuous-observation policy representation is introduced using Stochastic Kernel-based Finite State Automata (SK-FSAs). An SK-FSA search algorithm titled Entropy-based Policy Search using Continuous Kernel Observations (EPSCKO) is introduced and applied to the first ever continuous-observation Dec-POMDP/Dec-POSMDP domain, where it significantly outperforms state-of-the-art discrete approaches. This methodology is equally applicable to Dec-POMDPs and Dec-POSMDPs, though the empirical analysis presented focuses on Dec-POSMDPs due to their higher scalability. To improve convergence, an entropy injection policy search acceleration approach for both continuous and discrete observation cases is also developed and shown to improve convergence rates without degrading policy quality.

  • Open Access

    Semantic-level Decentralized Multi-Robot Decision-Making using Probabilistic Macro-Observations

    by Shayegan Omidshafiei, Shih-Yuan Liu, Michael Everett, Brett T. Lopez, Christopher Amato, Miao Liu, Jonathan P. How, John Vian

    Robust environment perception is essential for decision-making on robots operating in complex domains. Intelligent task execution requires principled treatment of uncertainty sources in a robot's observation model. This is important not only for low-level observations (e.g., accelerometer data), but also for high-level observations such as semantic object labels. This paper formalizes the concept of macro-observations in Decentralized Partially Observable Semi-Markov Decision Processes (Dec-POSMDPs), allowing scalable semantic-level multi-robot decision making. A hierarchical Bayesian approach is used to model noise statistics of low-level classifier outputs, while simultaneously allowing sharing of domain noise characteristics between classes. Classification accuracy of the proposed macro-observation scheme, called Hierarchical Bayesian Noise Inference (HBNI), is shown to exceed existing methods. The macro-observation scheme is then integrated into a Dec-POSMDP planner, with hardware experiments running onboard a team of dynamic quadrotors in a challenging domain where noise-agnostic filtering fails. To the best of our knowledge, this is the first demonstration of a real-time, convolutional neural net-based classification framework running fully onboard a team of quadrotors in a multi-robot decision-making domain.

  • Multi Source Data Integration for Aircraft Health Management

    by Estefan M. Ortiz, Ashish Babbar, Vassilis L. Syrmos, Gregory J. Clark, John L. Vian, Michael M. Arita

    Modern aircraft are capable of generating and collecting massive quantities of data from flight recorders, maintenance reports, logistics, and mission-readiness reporting systems. Current aircraft system health management schemes are developed based on data sources consisting of the aircraft's system operational conditions or maintenance and repair actions; however, crucial information regarding flight condition and situational parameters are often disregarded or used in a limited sense for diagnostics due to data accessibility issues or what can be an overwhelming volume of time-series data that is collected from modern onboard aircraft data recorders. With improved access to multiple sources of aircraft data, diagnostics and prognostics capabilities may be improved allowing flight crews, engineering, maintenance, and logistics users of an interoperable data system to make more informed decisions. Improved knowledge of the actual aircraft condition, usage and component life monitoring can significantly reduce aircraft operations and support (O&S) costs. In addition, integrating multiple sources of aircraft data can reduce total ownership cost (TOC), enhance mission safety and improve aircraft performance. There is a need to develop a multi-source data integration architecture for analyzing vast quantities of disparate data obtained from both onboard and off-board data sources. The multi-source data integration will be able to accurately combine disparate data and establish interconnectivity between distinct data sources to significantly improve data interoperability across multiple data sources. This type of information fusion can serve as the basis for developing advanced diagnostic and prognostic algorithms for legacy, current, and next- generation commercial and military aircraft.


  • Embedding Health Management into Mission Tasking for UAV Teams

    by Mario Valenti, Brett Bethke, Jonathan P. How, Daniela Pucci de Farias, John Vian

    Coordinated multi-vehicle autonomous systems can provide incredible functionality, but off-nominal conditions and degraded system components can render this capability ineffective. This paper presents techniques to improve mission-level functional reliability through better system self-awareness and adaptive mission planning. In particular, we extend the traditional definition of health management, which has historically referred to the process of actively monitoring and managing vehicle sub-systems (e.g., avionics) in the event of component failures, to the context of multiple vehicle operations and autonomous multi-agent teams. In this case, health management information about each mission system component is used to improve the mission system's self-awareness and adapt vehicle, guidance, task and mission plans. This paper presents the theoretical foundations of our approach and recent experimental results on a new UAV testbed.


  • The MIT Indoor Multi-Vehicle Flight Testbed

    by Mario Valenti, Brett Bethke, Daniel Dale, Adrian Frank, James McGrew, Spencer Ahrens, Jonathan P. How, John Vian

    This paper and video present the components and flight tests of an indoor, multi-vehicle testbed that was developed to study long duration UAV missions in a controlled environment. This testbed is designed to use real hardware to examine research questions related to single- and multi-vehicle health management, such as vehicle failures, refueling, and maintenance. The testbed has both aerial and ground vehicles that operate autonomously in a large, indoor flight test area and can be used to execute many different mission scenarios. The success of this testbed is largely related to our choice of vehicles, sensors, and the system's command and control architecture. The video presents flight test results from single- and multi-vehicle experiments over the past year.


  • Health Management Engineering Environment and Open Integration Platform

    by Kirby Keller, Andrew Baldwin, Stan Ofsthun, Kevin Swearingen, John Vian, Tim Wilmering, Zachary Williams

    The Phantom Works' role at The Boeing Company is to facilitate the development and transition of emerging technologies into Boeing products to provide a competitive advantage. Phantom works is developing a health management engineering environment (HMEE) to foster and transition health management technology from industry, academia and government technology base to Boeing Products. The HMEE consists of a program analysis and modeling environment, a development environment and an operations environment. The Program Analysis and Modeling Environment provides the processes and tools to do performance/cost driver analysis, root cause analysis, solution formulation, and cost/benefit trade studies including fleet simulation modeling. The development environment provides the design processes and synthesis tools required to develop solutions. The operations environment provides the processes and tools to integrate, test and mature the hardware and software elements of the health management solution. Major elements of the HMEE are an open software reference architecture, access to hardware laboratories to characterize degraded components/ subsystems, a data repository to store and access laboratory and field data, and hardware and software to support end to end technology demonstrations. It will support progressive levels of integration and demonstration from a single technology on a PC to the integration of this technology into a complete IVHM system with hardware in the loop as needed to provide vehicle data and address integration into vehicle avionics or a ground support system. Significant elements of the HMEE are in place with additional expansion and integration ongoing. The Boeing Phantom Works IVHM team is acquiring hardware, tools and algorithms from a number of suppliers to create a pool of resources for system level, end to end demonstrations of IVHM applications and development tools.


  • Multi-platform Airplane Health Management

    by Greg J. Clark, John L. Vian, Michael E. West, Vassilis L. Syrmos, William F. Randolph, William J. Hardman, Matthew A. Shawver, Geoff J. Hanson

    The Integrated Airplane Health Management (IAHM) program is a Navy research activity led by The Boeing Company with support from the University of Hawaii, Referentia Systems Incorporated, and Impact Technologies, LLC. The program focuses on interoperable multi-platform aircraft health management data handling and analysis methods applicable to military aircraft and commercial air transportation. A primary goal of the program is to significantly increase aircraft system reliability, safety, maintainability, availability, and affordability, resulting in improved mission performance and operational capabilities. As such, improved aircraft data accessibility and interoperability is viewed as a key enabler. The program consists of three phases leading to a proof-of-concept demonstration to show technical feasibility and potential payoffs for both military aircraft and commercial airlines. Software data loading translators have been developed and are being used to populate a new MIMOSA-based multi-platform health management database that can interface with common algorithms to support condition-based maintenance practices. In particular, this paper provides an overview of the concept and the efforts of the IAHM team in accommodating a variety of representative aircraft health data from F/A-18, C-17, and 737 platforms.12 A graphical user interface has been developed to access the IAHM database for data visualization, new diagnostic and prognostic algorithm research and development, and also for processing new datasets with the health management algorithms being developed. The resulting IAHM database schema, data interface methods, and algorithms are applicable to new, derivative, and current aircraft.


  • Stochastic Optimal Control of a Servo Motor with a Lifetime Constraint

    by Alexander Bogdanov, Stephen Chiu, Levent U. Gokdere, John Vian

    We consider a linear quadratic optimal regulation problem of a servo motor in the presence of stochastic load disturbance subject to a constraint that establishes a desired motor winding lifetime. To satisfy the constraint, a family of LQR control designs is parameterized with a single scalar performance weight that establishes a trade-off between performance and control power. Power analysis approach is then used to find the optimal value of the parameter that provides maximum disturbance rejection control in the given LQR family subject to the desired motor lifetime constraint