Cambridge, MA, United States

Charles River Analytics Inc.

www.cra.com
Cambridge, MA, United States
SEARCH FILTERS
Time filter
Source Type

Grant
Agency: Department of Transportation | Branch: | Program: SBIR | Phase: Phase I | Award Amount: 149.99K | Year: 2016

The connected vehicles program is a multimodal U.S. DOT initiative that uses a dedicated wireless communication technology to enable safer, smarter, and greener surface transportation. While significant efforts are being made to bring motor vehicles and transportation infrastructure onto this connected network, bicycles have been largely overlooked. As a result, a significant need exists to bring cyclists onto this network, both to enable other connected vehicles and infrastructure to be made aware of their presence, and to enable cyclists to take advantage of the safety and transportation benefits available when receiving information from other connected entities. To address this challenge, we are proposing to design and demonstrate a Multimodal Alerting Interface with Networked Short‐range Transmissions (MAIN‐ST). MAIN‐ST will bring cyclists onto the vehicle to vehicle (V2V) and vehicle to infrastructure (V2I) networks using a hardware‐agnostic approach, whereby a commercially available DSRC radio and optionally deployable state‐sensing hardware can be deployed to generate a robust basic safety message for bicycles (BSM‐B). In addition, MAIN‐ST allows cyclists to take advantage of being on these networks by providing automated hazard assessment capabilities with a multimodal alerting symbology designed to communicate hazard information to cyclists in an intuitive and non‐distracting format.


Grant
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase II | Award Amount: 749.96K | Year: 2015

Current mission planning interfaces are difficult to understand and cumbersome to use, resulting in few operators utilizing the full power of advanced planning systems. To address this shortfall, we will design and demonstrate Intuitive User Interfaces for Task-Tailored planning (INTUIT). These interfaces will increase the usability of planning systems and the efficiency of operators with varying skill levels across a range of vehicles, mission contexts, and unique tasks. INTUIT will provide targeted support to novice and expert mission planners by adapting to unique operator, task, and mission needs, as well as fluidly exposing opportunities for advanced planning functionality when appropriate. Under the proposed Phase II INTUIT effort, we will first expand prior analyses and workflow models to characterize specific capabilities, tasks, and workflows of mission planners. We will then leverage these analyses and models to refine our design of INTUIT displays for mission planning activities. Finally, we will expand the breadth and depth of our current INTUIT prototype. We will use these prototypes to demonstrate and evaluate our approaches and deeply explore the complexities of mission planning across vehicles, operators, and mission contexts.


Grant
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase II | Award Amount: 749.98K | Year: 2016

NASA missions include long periods of low workload followed by sudden high-tempo operations, a pattern that can be detrimental to situational awareness and operational readiness. An unobtrusive system to measure, assess, and predict Astronaut cognitive workload can indicate when steps should be taken to augment cognitive readiness. This system can also support testing and engineering (T&E); engineers can accurately evaluate the cognitive demands of new tools and systems, as well as how they affect task performance. In our Phase I effort, Charles River Analytics designed and demonstrated a system for Cognitive Assessment and Prediction to Promote Individualized Capability Augmentation and Reduce Decrement (CAPT PICARD). CAPT PICARD: (1) robustly and unobtrusively performs real-time synchronous data collection with a suite of sensors to provide a holistic assessment of the Astronaut; (2) extracts, fuses, and interprets the best combination of indicators of Astronaut state; (3) comprehensively predicts performance deficits, optimizing the likelihood of mission success; and (4) displays the data to support the information requirements of any user. The solicitation defined the following Phase I goals: review physiological, neurophysiological, and cognitive assessments in extreme environments and long duration missions; design an algorithm to assess workload. We did focus on these goals; however, we went beyond them to also demonstrate a functional prototype by the end of Phase I. Based on the success of this Phase I effort, we recommend a Phase II effort to refine and develop each component of CAPT PICARD, and iteratively evaluate this system in an undergraduate lab, at a T&E lab at Johnson Space Center (JSC), and in a mission-like analog environment at JSC. Successful completion of these tasks will result in a tool that can both dramatically improve Astronaut mission readiness and the design and development of tools Astronauts use to carry out mission objectives.


Grant
Agency: Department of Defense | Branch: Defense Advanced Research Projects Agency | Program: SBIR | Phase: Phase II | Award Amount: 1.50M | Year: 2015

Model-based programming (MBP) languages such as SysML and RMPL provide declarative descriptions of the structure, processes, functions, and context of a system. They are particularly useful for cyber-physical systems that interact with the environment through sensors and actuators. MBP languages are beneficial both in the design of systems and in the control of systems in real time. However, existing MBP languages are deterministic and do not model uncertainties that influence the systems behavior. We propose to develop Probabilistic Model-Based Programming Techniques for Prediction, Analysis and Control (PROMPT). PROMPT will provide probabilistic extensions to SysML, enabling that to enable the representation of uncertainties in both the structure and the behaviors of a complex system, as well as interactions between the structure and behaviors. PROMPT will provide inference services for predicting the behavior of a system under uncertainty and estimating the current state of the system from noisy sensors. These inference services will be used to predict performance and analyze models at design time and to control a cyber-physical system at runtime based on probabilistic beliefs about the state of the system. PROMPT will be applied to and evaluated on a significant cyber-physical system such as an unmanned undersea vehicle.


Grant
Agency: Department of Defense | Branch: Defense Health Program | Program: SBIR | Phase: Phase II | Award Amount: 999.96K | Year: 2015

US military medical personnel may be deployed to a variety of operational environments where their success saving lives depends on their ability to act quickly and effectively, both as individuals and as teams. Therefore, effective training must go beyond individual skills to include interactions among team members, and how those interactions transfer to operational environments. Currently, trainers must infer competence by observation alonea challenging task. Automatically sensing indicators of cognitive workload can augment performance observations, offering insight into factors underlying that performance. To address this issue, Charles River Analytics has designed and demonstrated the feasibility of a system to augment training by Monitoring, Extracting, and Decoding Indicators of Cognitive workload (MEDIC). Building on that success, we now propose a Phase II effort to iteratively evaluate and refine MEDIC, enabling the disambiguation of potential cognitive workload indicators from other causes, such as physical exertion. Our effort combines: (1) a multimodal suite of unobtrusive, field-ready neurophysiological, physiological, and behavioral sensors; (2) complex event processing to extract and fuse the best indicators of cognitive workload and team dynamics from the multiple, high-volume data streams originating from the sensor suite; and (3) probabilistic modeling techniques to interpret those indicators for easy understanding.


Grant
Agency: Department of Defense | Branch: Air Force | Program: SBIR | Phase: Phase II | Award Amount: 750.00K | Year: 2015

ABSTRACT: New 4.5th and 5th-generation USAF fighter aircraft (e.g., the F-22 and F-35) can be adapted for multiple missions and offer a wealth of emerging technological capabilities. In these aircraft, pilots must master new tactics and procedures that differ significantly in comparison to previous generation aircraft. However, comprehensively training pilots across such emerging tactics and procedures is not economical in live exercises. To effectively, efficiently, and economically prepare pilots for next-generation multirole aircraft, we conducted a Phase I effort to design and demonstrate a Next-Generation, Multirole Fighter Instruction and Rehearsal Environment (GeMFIRE). In Phase I, we successfully established the feasibility of our approach, integrating a variety of low-cost, commercial devices. We demonstrated a DIS-enabled software simulation environment for simultaneous multi-pilot missions, and a scenario creation tool for multi-mission exercise design. Based on the successful completion of Phase I, we propose a Phase II effort to design, develop, and evaluate GeMFIRE. In Phase II, we will: (1) refine our work analysis with 4.5/5th-generation aircraft experts; (2) provide a wider range of displays and controls in our adaptive training environment; (3) extend probabilistic training models within our scenario authoring tool; (4) integrate these components into a suite of NICE-compatible flight simulation solutions. BENEFIT: The primary benefit of GeMFIRE will be in supporting cost-effective training and rehearsal for next-generation, multirole aircraft pilot squadrons across a spectrum of low- to medium-fidelity integrated simulator solutions. However, we recognize considerable benefit in coordinating with additional aircraft (e.g., F-22s, Unmanned Aerial Vehicles) and ground forces (e.g., air operations centers), particularly those with multiple mission sets (e.g., ISR, strike). We will also extend our commercial agent development environment, AgentWorks, with authoring capabilities that will enhance AgentWorks appeal in the Serious Games and Live, Virtual, and Constructive (LVC) Simulation training markets (e.g., introducing more intelligent adversary behaviors). Our intelligent agent models coupled with robust displays and control devices will also result in opportunities in emerging home entertainment markets (e.g., immersive video games).


Grant
Agency: Department of Defense | Branch: Air Force | Program: SBIR | Phase: Phase II | Award Amount: 750.00K | Year: 2015

ABSTRACT: Maintaining the efficacy of military plans is difficult, yet critical. A plan may become irrelevant because the world has diverged from the plans assumptions; if there are changes to many related plans, they may become incoherent, unable to generate well-synchronized and coordinated operations. The Adaptive Planning process partly addresses this problem through cyclical, collaborative plan review and maintenance. However, to fully address this problem, the military needs a highly-automated and effective system that continuously maintains living plans in response to rapidly changing conditions. Charles River Analytics proposes to implement and evaluate a system for plan Coherence and Relevance Monitoring and Adjustment (CARMA) to provide an affordable, open, and extensible solution to help generate and maintain living plans. To create this system, we will develop: (1) a sophisticated catalog that represents plan elements (e.g., goals, assets, terrain) and their relationships (e.g., supported, supporting, and collateral) across time and space; (2) monitoring and analysis algorithms that can automatically guide Warfighters to adapt plans to changes that threaten plan coherence or relevance; and (3) an information extraction and routing service that grounds the living plan process in accurate, complete, and timely information. BENEFIT: The research performed under this effort will have immediate benefit to the Air Operations Center Weapon System (AOC WS) across multiple regional and functional AOCs. Specifically, it will increase the productivity of current planners as it allows them to attend to the most significant implications of changes in the operating environment, thereby increasing plan quality as planners spend more time thinking creatively and collaboratively. It will also improve force effectiveness as forces will go into battle with plans that achieve effective unity of effort. These benefits will apply across air, maritime, ground, and space domains. This research will also enhance our commercial AgentWorks software development kit for deploying hybrid computational reasoning into an enterprise.


Grant
Agency: Department of Defense | Branch: Air Force | Program: SBIR | Phase: Phase II | Award Amount: 749.98K | Year: 2015

ABSTRACT: Simulations supporting Close Air Support (CAS) training and the Joint Terminal Attack Controller (JTAC) training are laborious to configure and expensive to manage with live personnel, which results in training that is limited in scope. For example, current simulations, like the Indirect Fire Forward Observer Trainer (I-FACT), require numerous observer/controllers to participate as White Force players in the simulation, driving up the manpower cost of conducting JTAC training. To fulfill the roles, communications, and behaviors of White Force participants in simulation-based training environments, we conducted a Phase I effort to design and demonstrate a full-scope Constructive Logic and Behavior Engine for Efficient and Responsive Training (COLBERT). In Phase I, we successfully established the feasibility of our approach by providing a graphical development and run-time environment to construct and execute CAS-related behavior models. COLBERT integrated our in-house, intelligent agent platform AgentWorks with the popular video game engine Unity, and included experience management tools that support Trainers across the multiple phases of training development. Based on the successful completion of Phase I, we propose a Phase II effort to design, develop, and evaluate COLBERT. In Phase II, we will deliver an initial set of simulation-based tools to support JTAC training. BENEFIT: COLBERT will provide immediate and tangible benefits to training developers by supporting effective and cost-efficient training across a wide range of domains and simulation environments, including the Joint Terminal Attack Controller Training Rehearsal System being developed at AFRL. COLBERT will also provide considerable benefit for other simulation-based training, including flight and space maintenance simulation training. In addition, we plan to enhance our user-friendly agent development environment, AgentWorks, with training capabilities that will increase its appeal in both the simulation-based training and entertainment (e.g., video games) markets.


Grant
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase II | Award Amount: 999.92K | Year: 2015

Despite advances in automated instruction, virtual environment (VE)-based shiphandling trainers still significantly depend upon expert mariners, with individual instructors typically working with only a single active student at a time. This bottleneck makes it difficult to realize the full benefit of VE training systems. Promising support technologies, such as the Conning Officer Virtual Environment (COVE) Intelligent Tutoring System (ITS) and the CRESST Assessment Engine (CAE), offload many manual training tasks, but also introduce new supervisory challenges to successfully monitor the progress of multiple students. To address these challenges and enhance the overall efficiency of COVE training, we will extend the design, development, and evaluation of a Shiphandling Educator Assistant for Managing Assessments in Training Environments (SEAMATE). SEAMATE will enable smaller numbers of instructors to supervise larger groups of students through a combination of improved student performance tracking, event-based automated feedback, and efficient, contextualized alerting. SEAMATEs instructor-oriented user interface (UI) will support at-a-glance awareness of multiple students performance, providing dynamic feedback to direct the instructors attention and improve awareness of individual student progress and intervention needs. Overall, SEAMATE will increase the effectiveness and efficiency of VE-based training methods, enabling highly trained shiphandling instructors to work with larger training cohorts.


Grant
Agency: Department of Defense | Branch: Missile Defense Agency | Program: STTR | Phase: Phase II | Award Amount: 998.13K | Year: 2015

In our Adaptive Management and Mitigation of Uncertainty in Fusion (AMMUF) project, we will model the entire multi-sensor fusion process as a probabilistic model and reason about the different design and algorithmic decisions that can be made by system engineers. This fusion model will use standard fusion system representations and ideas from statistical relational learning field to create flexible and expandable fusion systems. AMMUF will enable system engineers to determine the optimal fusion configuration in different missile defense contexts, giving battlefield operators the most accurate and efficient information about missile threats. Approved for Public Release, 15-MDA-8303 (1 July 15)

Loading Charles River Analytics Inc. collaborators
Loading Charles River Analytics Inc. collaborators