This proposal seeks funding for the Center for Safety, Security, and Rescue Research studies conducted by the University of Minnesota (lead) and its research partner at the University of Denver. Funding Requests for Fundamental Research are authorized by an NSF approved solicitation, NSF 10-507. The solicitation invites I/UCRCs to submit proposals for support of industry-defined fundamental research.
The proposal is about further developing ongoing robot design activities for safety and security. The presented research focus is in the area of underwater robot design, and it describes activities regarding the waterproofing, buoyancy control, camera stabilization and sensor fusion strategies. The proposed work builds on the strong previous track record, and the team is strong. Unlike most existing approaches, the proposed effort should provide miniature size robots with high mobility.
This proposal presents a set of robotic systems (including an amphibious, tumbling robot) and algorithms that capitalize on earlier pertinent work at the Industry/University Cooperative Research Center for Safety, Security, and Rescue Research. The proposal also offers innovative stabilization algorithms and hardware for the sensor suite in order to improve the quality of data provided to the first responder. In addition, innovative sensor fusion schemes for underwater navigation are proposed. The broader impacts of the proposed work range from a planned inclusion of undergraduates into the research plan to the possibility that life-saving commercial devices could eventually result from the work described. The connection of this work with the NSF Safety, Security, and Rescue Research Center will ensure that the proposed technologies will be exposed to first responders at the proposed field trials.
The main outcomes of this project are: 1- Development of a video stabilization algorithms that can be used for robotic and non-robotic applications. Video stabilization removes unwanted motion from video sequences, often caused by vibrations or other instabilities. This improves video viewability and can aid in detection and tracking in computer vision algorithms. We have developed a digital video stabilization process using scale-invariant feature transform (SIFT) features for tracking motion between frames. These features provide information about location and orientation in each frame.The orientation information is generally not available with other features, so we employ this knowledge directly in motion estimation. We use a fuzzy clustering scheme to separate the SIFT features representing camera motion from those representing the motion of moving objects in the scene. Each frame’s translation and rotation is accumulated over time, and a Kalman filter is applied to estimate the desired motion. 2- Development of a FPGA-based reconfigurable hardware system for research in robotics. This project focuses on a challenging problem: design a reconfiguration architecture to support visual servoing application. This architecture, termed "Morphing Crossbar", is a hybrid mechanism that combines the low resource consumption of bus-based Partial Dynamic Reconfiguration (PDR) mechanisms with the speed of a point-to-point crossbar. The abstraction of the hardware module interface to the re-routing design pattern employed reduces consumption of internal resources, such as hex and long lines, and enhances hardware reusability. Second, the logical separation of each module into a functional part and an interconnect relation part allows the PDR execution latency to remain low. Experimental results on a "visual servoing" task demonstrates the Morphing Crossbar utilizes, on average, 72% fewer routing resources to interconnect modules, and performs 3.6 times faster in PDR compared to other methods.
