tactical aerial resupply vehicle
play

Tactical Aerial Resupply Vehicle Shawn Recker, Rob Baltrusch, Joe - PowerPoint PPT Presentation

Autonomous Capabilities of the Joint Tactical Aerial Resupply Vehicle Shawn Recker, Rob Baltrusch, Joe Rosenthall, Jon Worobey, Christiaan Gribble, and Mark Butkiewicz SURVICE Engineering Company Applied Technology Operation 4603A Compass


  1. Autonomous Capabilities of the Joint Tactical Aerial Resupply Vehicle Shawn Recker, Rob Baltrusch, Joe Rosenthall, Jon Worobey, Christiaan Gribble, and Mark Butkiewicz SURVICE Engineering Company Applied Technology Operation 4603A Compass Point Road, Belcamp, MD 21017 shawn.recker@survice.com SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  2. OUR PEOPLE With a staff of approximately people, SURVICE employees have backgrounds in engineering, physics, mathematics, chemistry, computer science, acquisition, technical writing, training, and other technical and administrative fields. And many of our personnel have DoD and/or operational military experience. Using a matrix approach, we draw on expertise from across our workforce to build customized work teams for each project and customer. The result is a successful, skilled workforce and many satisfied customers. 2 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  3. ABOUT SURVICE Since 1981, SURVICE has supported the DoD community, as well as the homeland security, advanced technologies, environmental, and commercial markets. We have also worked hard to retain the original vision upon which the company was founded — to provide the warfighter with the safest, most survivable, and most effective equipment possible. Our customer-acclaimed support primarily falls into five core business areas: Test & Evaluation Modeling & Dimensional Studies & Information Simulation/ Metrology and Analysis Technologies & Providing experience Software Reverse Management Conducting a wide in the test and Engineering Engineering variety of safety, Collecting, maintaining, evaluation roles that Delivering a broad Offering customers distributing, storing, survivability, are vital to the design, range of modeling and precision dimensional verifying, and effectiveness, and acquisition, and simulation, software metrology, modeling, protecting the nation’s systems engineering fielding of all new and engineering, and and related services. studies. critical defense modified combat systems engineering information. systems. capabilities. 3 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  4. Partnership 4 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  5. Joint Tactical Aerial Resupply Vehicle (JTARV) 5 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  6. Target Platform Architecture Flight Controller Common Control Radio Module Transceiver Ground Control Radio Controller Station 6 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  7. Ground Control Station 7 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  8. Common Control Module 8 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  9. Autonomous Landing 9 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  10. Autonomous Landing Pipeline Incoming Image DNN Detection Photogrammetry #39 Craft Movement Target Direction Disarm Issued Computed 10 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  11. Pipeline Motivation 11 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  12. Underlying Technologies Real-Time In-Situ Advanced 3D Intelligent Video Reconstruction Analytics 12 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  13. DetectNet Output Error FCN Input NVIDIA Parallel Forall: DetectNet [1] 13 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  14. DetectNet Dataset Creation 14 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  15. DetectNet Training with DIGITS 15 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  16. HawkEye™ Advanced 3D Reconstruction Technology • Project goals: • Low-cost means of capturing accurate 3D data in the field for post-IED events in a secure environment • Data to feed analysis tools to estimate IED event and post-event structural integrity • SURVICE developing HawkEye to combine: • Structure-from-Motion (SfM) Technology • Image analysis can be done either on the device or in a secure computing environment • Photogrammetry Technology • Hybrid approach substantially improves accuracy of SfM from imagery alone • Low-cost 3D Scanning Technology • Increases 3D point cloud data and accuracy, especially for large featureless surfaces 16 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  17. HawkEye™ Photogrammetry Technology • SURVICE has developed hybrid photogrammetry technology for SfM to significantly improve accuracy over current SfM methods • Achieving an order-of-magnitude improvement in 3D resolution/accuracy • Approach & preliminary results published at CMSC 2014 conference CMSC 2014 Paper Recently Published & Presented by SURVICE and UC Davis on Hybrid SfM Methodology SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES 17

  18. HawkEye™ SfM Technology • SURVICE & UC Davis developing methods to produce dense reconstructions given initial sparse estimate • Currently uses parallax paths constraints imposed by path of moving camera • Working to relax constraints to create a more flexible dense reconstruction pipeline HawkEye Dense SfM Existing State-of-the-Art Joint Paper by UC Davis and SURVICE on Dense SfM Methodology 18 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  19. Future Applications: Dense Aerial Reconstruction Parallax-Path Based Dense Reconstruction [3] 19 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  20. Future Applications: Real-Time UAV Target Tracking 20 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  21. Acknowledgement This work was supported in part by the US Marine Corps SBIR program, the US Air Force SBIR program, and SURVICE Engineering’s Internal R&D program. The authors thank their colleagues in the Applied Technology Operation at SURVICE Engineering for their support. 21 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  22. References [1] DetectNet: Deep Neural Network for Object Detection in DIGITS Andrew Tao, Jon Barker, and Sryia Sarathy https://devblogs.nvidia.com/parallelforall/detectnet-deep-neural-network-object-detection-digits/ [2] Depth Data Assisted Structure-from-Motion Parameter Optimization and Feature Track Correction Shawn Recker, Christiaan Gribble, Mikhail M. Shashkov, Mario Yepez, Mauricio Hess-Flores, and Kenneth I. Joy Applied Imagery Pattern Recognition (AIPR), IEEE 2014 [3] Efficient Dense Reconstruction Using Geometry and Image Consistency Constraints Mikhail M. Shashkov, Jason Mak, Shawn Recker, Connie Nguyen, John Owens, Kenneth I. Joy AIPR 2015: Applied Imagery Pattern Recognition Workshop, Washington DC, 2015. 22 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  23. Questions 23 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

  24. Providing unparalleled services and technology solutions in support of National Defense and the U.S. Warfighter. Serve our customers with specialized and innovative solutions, while promoting responsible growth and maintaining focus on our employees, values, and ethical standards. 24 SMALL COMPANY VALUES, BIG COMPANY CAPABILITIES

Recommend


More recommend