Agency: National Aeronautics and Space Administration | Branch: | Program: STTR | Phase: Phase II | Award Amount: 749.95K | Year: 2014
SSCI and MIT propose to further develop, implement and test the Integrated Mission Planning & Autonomous Control Technology (IMPACT) system software for autonomous ISR missions employing collaborating UAVs. IMPACT system is based on real-time learning about dynamic and stochastic environments, and on a capability to autonomously react to contingencies while satisfying the mission objectives and the overall flight safety. Phase II focus will be on real-time vehicle assignment & trajectory planning technologies for forest fire monitoring, overall system integration, and evaluation of its performance through computer and hardware-in-the-loop simulations and flight tests at Olin College or Great Dismal Swamp. Key technologies to be further developed & tested in Phase II include: (i) Vehicle assignment & real time trajectory generation for collaborative ISR for fire boundary identification using the MOTOR system (Multi-objective Trajectory Optimization & Re-planning); (ii) Robust on-line learning for prediction of the fire spread using the intelligent Cooperative Control Architecture (iCCA); (iii) Collaborative assignment for fire perimeter tracking with reactive trajectory planning based on predicted fire spread using MOTOR and iCCA; (iv) Contingency management, including the loss of vehicle, vehicle replacement & mitigation of lost communication link; and (v) Predictive camera pointing control based on predicted fire spread. The project will leverage a number of technologies recently developed by SSCI and MIT, and integrate various system modules within a flexible and user-friendly software product. Phase II deliverables will include the IMPACT software and accompanying documentation, while Phase III will be focused on commercialization of the IMPACT software.
Agency: Department of Defense | Branch: Defense Advanced Research Projects Agency | Program: SBIR | Phase: Phase I | Award Amount: 149.93K | Year: 2015
Currently small unmanned air vehicles (UAVs) are limited in their ability to autonomously fly in complex, cluttered, environments. This significantly limits the utility of small tactical UAVs to the warfighter/squad who must manually control the vehicle. Small UAVs require: a) scalable autonomous survivial behaviors to avoid collision with environmental structure and moving objects; b) Low size, weight, power, and cost. The SSCI SA3 system (Small Autonomous, Sensor Agnostic, Sense and Avoid) addresses these two challenges through bio-inspired reactive controller and visual threat-estimation algorithms. SA3 is built around a Steering Field which fuses obstacle representations from sensors of multiple modalities and combines them with the current navigation waypoint to generate reactive steering commands, avoiding obstacles while navigating. In the current program we will demonstrate SA3 with visual Expansion Rate technologies for the detection of self-moving objects (other aircraft, vehicles, people, etc) and environmental structure (walls, the ground, etc). In Phase I we will design the SA3 system architecture, perform a limited data collect, and demonstrate how key system components enable unassisted airborne navigation (take-off, waypoint traversal, landing). In Phase II, SSCI will work with one of our existing UAV partners to demonstrate the SA3 system in flight on a small UAV.
Agency: Department of Defense | Branch: Defense Advanced Research Projects Agency | Program: SBIR | Phase: Phase II | Award Amount: 1.50M | Year: 2015
We propose a Phase II effort to extend the game-theoretic technology to apply to the DARPA Distributed Battle Management (DBM) program. Our proposed AMDP (Adversarial Modeling for DBM Planning) system augments DBMs capabilities with a predictive model of adversarial uncertainty and action. We seek to answer the question How can we predict enemy responses and positions, and update those predictions based on incoming data AMDP uses models of various aspects of enemy COAs including decoys and mobile units and various related uncertainties including: enemy positions, asset types, and available actions in a SEAD-type (Suppression of Enemy Air Defenses) mission against ground targets. AMDP fuses prior, incoming, and predicted world knowledge to generate courses of action that maximize informational gain metrics, and thus minimize uncertainty, in order to ensure correct actions and protect against worst-case outcomes due to unexpected events. The targeted end result is a simulation assessment showing improved mission efficacy (measured by speed and accuracy in identifying and engaging true targets in a field of decoys) in high-uncertainty scenarios, within a third-party air battle simulation environment such as AFSIM or SIMAF.
Agency: Department of Defense | Branch: Defense Advanced Research Projects Agency | Program: SBIR | Phase: Phase II | Award Amount: 986.09K | Year: 2013
SSCI proposes a retooling of the Intelligent Course of Action Learning System (iCOALS) that leverages the important results from the Phase I effort and remedies the deficiencies of a rule-based approach when applied to complex Pursuit Evasion Games (PEGs). This modification is critical given the utility of the proposed approach to DARPA's Anti-Submarine Warfare (ASW) Continuous Trail Unmanned Vessel (ACTUV) program. ACTUV's"Track-and-Trail"problem is specific, real world instance of a PEG where ACTUV attempts to maintain close in sensor contact over a period of weeks with a manned diesel-electric submarine. This problem can be formulated as a PEG where the unmanned pursuit vehicle (ACTUV: propulsive superiority) attempts to maintain proximity with the evader (submarine: intellectual superiority). SSCI's specific approach to this problem provides an online Adversarial Autonomy (AA) engine that employs a Forward Reachability Model (FRM) formulation of a PEG game. This is important because it is impossible to formulate a rule-set that can handle all challenges from an intelligent adversary especially one with access to external effectors (surface traffic, active/passive decoys, weapons systems, boarding parties, etc.). Game-theoretic approaches, like FRM, model key performance parameters of the pursuer (ACTUV) as well as modeling or measuring those of the evader. Once modeled, efficient computational search algorithms (alpha-beta) are employed to ensure optimal outcomes for the current engagement configuration in real-time. It should be noted that engagement-level optimality is precluded in a rule-based approach with a finite rule set. Upgrading iCOALS with online adversarial models will eliminate the possibility track-and-trail will be easily broken by an intelligent adversary who quickly discovers and repeatedly exploits rule-set loopholes.
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase II | Award Amount: 2.46M | Year: 2013
Autonomous landing of a Vertical Takeoff and Landing Unmanned Air Vehicle (VTUAV), such as Fire Scout, on the deck of a moving ship requires a precise and accurate ship-relative navigation solution. The existing automated landing system, consisting of ship-based radar equipment and airborne transponder equipment that utilizes round-trip Radio Frequency (RF) characteristics to determine relative position, is prone to component failure as well as unavailability of the required RF spectrum issues that may require abandoning an expensive asset at sea. Image-based Navigation for Shipboard Landing (ImageNav-SL) system can act as a backup or replacement for the existing radar -based solution and enables recovery in the aforementioned scenarios. ImageNav-SL system generates ship-relative navigation estimates for its onboard guidance and control system during the final approach and terminal landing phases. ImageNav-SL uses existing visual landing aids on the ship deck and does not require any ship-to-platform synchronous datalink or GPS. In Phase-I, ImageNav-SL was developed using advanced computer vision methods and its feasibility successfully demonstrated. Phase II will expand on the concept design to specify a detailed system HW/SW design and test a prototype implementation using data collects and real-time open-loop flight tests using Northrop Grumman Corporation (NGC) Fire Scout test assets.
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase II | Award Amount: 917.05K | Year: 2014
Autonomous landing of a fixed-wing Unmanned Air Vehicle on the deck of a moving ship requires a precise and accurate ship-relative navigation solution. Existing landing systems consisting of combinations of GPS, Radar, and other radio-based communications, are prone to component failure as well as unavailability of the required RF spectrum issues that may require abandoning an expensive asset at sea. The Image-based Navigation for Fixed-Wing Shipboard Landing (ImageNav-FW) system can act as a backup, replacement, or complementary solution and enables recovery in the aforementioned scenarios. The ImageNav-FW system generates ship-relative navigation estimates for its onboard guidance and control system during the final approach and terminal landing phases. ImageNav-FW uses existing visual landing aids on the ship deck and does not require any ship-to-platform synchronous datalink or GPS. In Phase-I, ImageNav-SL was developed for VTUAV using advanced computer vision methods. Phase II of that program is expanding on the concept design to specify a detailed system HW/SW design and test a prototype implementation using data collects and real-time open-loop flight tests. In the ImageNav FW Phase II, SSCI will apply the ImageNav-SL algorithms to fixed-wing landing and augment them with state estimation of Visual Landing Aids (VLA), e.g. IFLOLS, and landing lights.
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase I | Award Amount: 80.00K | Year: 2015
Autonomous landing of a fixed wing Unmanned Air Vehicle on the deck of a moving ship requires a fast initializing, precise, and accurate ship-relative navigation solution able that is capable of functioning in emissions control scenarios (EMCON). Current state-of-the-art landing systems rely on RF emissions and are therefore inoperable in EMCON or are subject to the availability of a specific RF spectrum. The proposed RAIN-LANDR system is designed to provide fully EMCON compliant, continuous, high integrity range and bearing data for unmanned or assisted approach and landing. RAIN-LANDR leverages and integrates aircraft fast initializing pose estimation and navigation filter modules developed by SSCI to provide high accuracy precision ship-relative navigation data. High intensity, visible light beacons placed on the ship extend the range of operations and speed-up pose estimation, ultimately contributing to fast filter convergence. The objective of the Phase I effort will be the development of the RAIN-LANDR system concept, validation in simulation studies using physics-based renderings with accurate light source modeling. The Phase I effort will provide a solid foundation for the implementation and demonstration of an integrated system in Phase II.
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase II | Award Amount: 749.79K | Year: 2015
The Variable Camber Continuous Trailing Edge Flap (VCCTEF) concept offers potential improvements in the aerodynamic efficiency of aircraft through real time wing shaping. NASA and Boeing have been studying the suitability of this concept to address the drag reduction problem in aircraft with reduced-stiffness wings. However, reduced stiffness may lead to wing flutter. In addition, displacements of VCCTEF control surfaces are limited and subject to highly nonlinear and time-varying constraints. Hence control design needs to solve a constrained multi-objective optimization problem. To address these challenges, in Phase I SSCI carried out initial development and testing of the Drag Identification and Reduction Technology (DIRECT). The DIRECT software estimates wing structural modes on-line and uses that information in a robust predictive controller design. Based on using Evolutionary Optimization and off-line analysis, DIRECT estimates wind disturbances on-line and selects optimal controller parameters from a table lookup to achieve on-line drag minimization. Building upon the successful Phase I development, in Phase II we propose to extend the DIRECT approach and evaluate its performance through high-fidelity simulations and wind-tunnel testing. Specific Phase II tasks include: (i) Test Phase I flutter suppression algorithms and PSC algorithms in a GTM simulation with flexible modes and VCCTEF control surfaces; (ii) Extend and enhance the drag minimization approach by developing innovative Performance Seeking Control (PSC) algorithms; and (iii) Compare the features of PSC and other available performance seeking control algorithms through wind-tunnel testing at the University of Washington Aerodynamics Laboratory (UWAL). Professor Eli Livne of University of Washington and Mr. James Urnes, Sr. will provide technical support under the project. Phase III will focus on commercialization of SYMPTOM software to manned aircraft and UAS.
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase II | Award Amount: 749.99K | Year: 2015
In Phase I of this project, SSCI carried out initial development of the Generalized Guidance, Navigation & Control Architecture for Reusable Development (GUARD). The resulting framework is applicable across different Autonomous Rendezvous and Docking (AR&D) domains, and enables further development and testing of reusable GN&C software for such applications. GUARD is based on the key functional requirements for GN&C software for AR&D, with special emphasis on the commonality across different domains of operation and unique implementation requirements for GN&C algorithms in such domains. Phase I accomplishments include: (i) Augmented the flight-test proven on-line trajectory optimization and control algorithm with a Fault Detection, Identification and Accommodation (FDIA) capability, (ii) Extended SSCI's Vision Based Navigation (VBN) algorithms, recently demonstrated for shipboard landing flight experiments, to achieve centimeter-level positioning accuracy for the AR&D implementations, and demonstrated its robustness to docking pattern variations; (iii) Carried out a detailed study of common GN&C functions for AR&D, and developed a conceptual solution for a user interface enabling agile reconfiguration of domain-specific information; and (iv) Carried out initial analysis of System-level Performance Metrics for AR&D missions to facilitate V&V of the overall integrated GN&C system. Phase II will demonstrate an enhanced prototype of the GUARD with integrated GNC/FDIA/VBN software that will make it reusable in three disparate AR&D system domains. Demonstrations will be in simulation and hardware tests as follows: orbital AR&D (in simulation), planetary rover docking with a habitat (evaluations at Olin College on R-Gator platform), and a quadrotor close-proximity operation mission (evaluations at UT Austin on quadrotor platform). Phase III will focus on commercialization of the GUARD software and its implementation to future NASA Space Exploration missions.
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase II | Award Amount: 749.89K | Year: 2015
SSCI has developed a suite of SAA tools and an analysis capability referred to as ASPECT (Automated System-level Performance Evaluation and Characterization Tool). ASPECT encapsulates our airspace encounter generator, sensor/tracker fusion algorithms, and prediction, threat assessment, and avoidance modules. It also provides both component-level and system-level analysis that is required for evaluating how well SAA sensors and software meet fundamental safety requirements for UAS in the NAS. ASPECT consists of MESSENGER (Multi-aircraft Encounter Scenario Generator), ASSIST (AsynchronouS Sensor fusIon SysTem), FORECAST (Fast On-line Prediction of Aircraft State Trajectories), and REACT (Rapid Encounter Avoidance & Conflict Resolution) modules. Initial versions of FORECAST and REACT were designed under related projects. Phase I developed the ASSIST (Asynchronous Sensor Fusion System) capability, which fuses combinations of SAA sensors such as GRB, ABR, camera, and Mode C transponder for localizing non-communicating threats. ASPECT was then used to analyze ASSIST's estimation accuracy, with the objective of achieving the precision of ADS-B and rejecting spurious/clutter tracks. Phase II will: (i) Expand and validate the underlying sensor models and demonstrate capability using flight test data generated at Olin College (Needham, MA), (ii) Extend our REACT system, and (iii) Carry out SAA system-level analyses using ASPECT to illustrate the relationship between sensor suite (hardware) selection, component SAA software modules, and achievable safety performance of the integrated system. The result of Phase II efforts will be a complete flow-down error and risk analysis framework, which constitutes a major step toward the integration of UAS into the National Airspace System. Phase II plans have been reviewed by NASA's UAS Traffic Management Program and AeroVironment (letters of support attached), who we anticipate to be one of our early transition partners.