Technical Approach:

To address this problem, we propose a scaled down urban environment. The scaled environment is surrounded by a 6 DOF gantry to which the sensor suite can be attached. By programming the gantry to mimic the aircraft flight dynamics, contr4ol algorithms can be tested against real sensor input. Testing and evaluation (T&E) can then be performed on control algorithms before flight. To verify and validate (V&V) results, and SR100 robotic helicopter can be flown in the full scale environment at the Piasecki Aircraft airfield.

Motivation:

The lack of fully integrated UAV testing in realistic mission conditions results in ad-hoc tests with mixed results. To address this issue, we seek to design a testing environment that accurately represents a real-world flight test. The goals of this project are to:

• Provide an environment for testing UAV mission
• Integrate vehicle hardware and software into testing
• The project goals are to assess the shared fate hypothesis and understand how motion cueing can improve UAV operation.
• Accurately simulate environmental conditions in a consistent and controlled manner

Success in these goals will provide the robotics community with a comprehensive method for designing UAV’s, opening up previously expensive/dangerous research directions.

Progress:

• The Piasecki airfield has been constructed at 1/87th scale
• A KLT feature tracker has been implemented showing that computer vision is a viable/scalable technology for testing
• Seminal work has been identified that can be replicated at small scale.

Publications:

Keith W. Sevcik and Paul Y. Oh, “Testing Unmanned Aerial Vehicle Missions in a Scaled Environment”, Journal of Intelligent Robot Systems, June 2008.

Contact: Prof. Paul Oh, Director, Drexel Autonomous Systems Lab, Email: paul@coe.drexel.edu