A5084117282- Robotics and Sensor-Based Localization
- Target Tracking and Data Fusion in Sensor Networks
- Advanced Vision and Imaging
- Robotic Path Planning Algorithms
- Inertial Sensor and Navigation
- Indoor and Outdoor Localization Technologies
- Advanced Image and Video Retrieval Techniques
- Distributed Sensor Networks and Detection Algorithms
- Distributed Control Multi-Agent Systems
- Video Coding and Compression Technologies
- Advanced Data Compression Techniques
- Fault Detection and Control Systems
- 3D Surveying and Cultural Heritage
- Satellite Image Processing and Photogrammetry
- GNSS positioning and interference
- Video Surveillance and Tracking Methods
- Underwater Vehicles and Communication Systems
- Image and Video Quality Assessment
- Aerospace Engineering and Control Systems
- UAV Applications and Optimization
- Embedded Systems Design Techniques
- Anomaly Detection Techniques and Applications
- Remote Sensing and LiDAR Applications
- Image and Video Stabilization
- Space Satellite Systems and Control
U.S. Air Force Institute of Technology
2012-2025
Wright-Patterson Air Force Base
2011-2024
California Institute of Technology
2023
United States Air Force Research Laboratory
2011-2018
Sensors (United States)
2018
University of Arizona
2016-2018
Australian Centre for Robotic Vision
2018
The University of Sydney
2018
Institute of Electrical and Electronics Engineers
2014-2016
Utah State University
2016
The unprecedented volume and rate of transient events that will be discovered by the Large Synoptic Survey Telescope (LSST) demands astronomical community update its followup paradigm. Alert-brokers -- automated software system to sift through, characterize, annotate prioritize for critical tools managing alert streams in LSST era. Arizona-NOAO Temporal Analysis Response Events System (ANTARES) is one such broker. In this work, we develop a machine learning pipeline characterize classify...
In this paper, we investigate the nonlinear observability properties of bearing-only cooperative localization. We establish a link between and graph that represents measurements communication robots. It is shown theoretic like connectivity existence path two nodes can be used to explain system. obtain maximum rank matrix without global information derive conditions under which achieved. Furthermore, show for complete observability, all in must have at least different landmarks known location.
This paper presents a method for localizing ground-based object when imaged from small fixed-wing unmanned aerial vehicle (UAV). Using the pixel location of target in an image, with measurements UAV position and attitude, camera pose angles, is localized world coordinates. study possible error sources localization sensitivities to each source. The has been implemented experimental results are presented demonstrating within 11 m its known
Training a convolutional neural network (CNN) for real‐world applications is challenging due to the requirement of high‐quality labeled imagery. This study employs pseudo‐labeling and transfer learning, built upon 6D pose estimation framework. A CNN trained on synthetic images predicts bounding boxes (bbox) an object's components in real image. With as few four bbox predictions, framework solves relative camera reprojects bboxes all onto that The reprojections allow filtering bad common...
This paper considers the problem of target handoff between Unmanned Aerial Vehicles (UAVs) in a GPS denied environment, and focuses on design evaluation an estimation strategy for determining relative pose aircraft. The approach presented this has three distinct components that act concert to achieve overall objective. First, novel cooperative control is used determine from IMU peer-to-peer ranging radio data without any priori knowledge either aircraft's pose. Next, measurement calculated...
This paper presents distributed formation control of a multi-agent system to encircle maneuvering target using Lyapunov and graph theories with emphasis on consensus cooperation. The proposed approach embeds algorithm into robust controller capture whose information is partially known. To address this issue, cooperative strategy in conjunction enable fleet UAVs either escort desired destination or keep the movement restricted certain set order gain superiority. We have employed sliding mode...
Abstract Once realized, autonomous aerial refueling will revolutionize unmanned aviation by removing current range and endurance limitations. Previous attempts at establishing vision-based solutions have come close but rely heavily on near perfect extrinsic camera calibrations that often change midflight. In this paper, we propose dual object detection, a technique overcomes such requirement transforming imagery directly into receiver aircraft reference frame probe-to-drogue vectors...
This paper discusses a computer vision algorithm and control law for obstacle avoidance small unmanned air vehicles using video camera as the primary sensor. Small UAVs are used low altitude surveillance ∞ights where unknown obstacles can be encountered. given capability to navigate in uncertain environments if identifled. presents an detection methodology feature tracking forward looking, onboard camera. Features found Harris Corner Detector tracked through multiple frames which provides...
Cooperative missions for Miniature Air Vehicles (MAVs) require accurate position, velocity, and attitude estimates all MAVs within the group its successful completion. This paper details a cooperative methodology MAV navigation in times of Global Positioning System (GPS) outages or GPS denied areas. In this method, each attitude, velocity sensor range, including itself. Each collects IMU measurements from neighboring fuses these with relative range bearing taken every range. collected data...
To enable navigation of miniature aerial vehicles (MAVs) with a low-quality inertial measurement unit (IMU), external sensors are typically fused the information generated by IMU. Most commercial systems for MAVs currently fuse GPS measurements IMU to navigate MAV. However there many scenarios in which an MAV might prove useful, but is not available (e.g., indoors, urban terrain, etc.). Therefore several approaches have recently been introduced that couple from visual (usually captured...
<h3>Abstract</h3> This tutorial presents the factor graph, a recently introduced estimation framework that is generalization of Kalman filter. An approach for constructing with its associated optimization problem and efficient sparse linear algebra formulation, described. A comparison filters presented, together examples generality graphs. brief survey previous applications graphs to navigation problems also presented. Source code extended filter generating in this paper available at...
To enable ubiquitous wireless multimedia communication, the bottlenecks to communicating data over channels must be addressed. Two significant which need overcome are bandwidth and energy consumption requirements for mobile communication. In this paper, we address dissipation by adapting image compression parameters current communication conditions constraints. We focus on JPEG algorithm, present results of varying some dissipation, required, quality received. a methodology selecting in...
As the technology sizes of semiconductor devices continue to decrease, effect nanometer technologies on interconnects, such as crosstalk glitches and timing variations, become more significant. In this paper, we study energy dissipation in interconnects. We propose a new power estimation technique which considers DSM effects, resulting significantly accurate estimates than transition-count based methods for on-chip also introduce an minimization attempts minimize large voltage swings across...
Accurate platform localization is an integral component of most robotic systems. As these systems become more ubiquitous, it necessary to develop robust state estimation algorithms that are able withstand novel and non-cooperative environments. When dealing with environments, little known a priori about the measurement error uncertainty, thus, there requirement uncertainty models algorithm be adaptive. Within this paper, we propose batch covariance technique, which enables through iterative...
In this paper, we investigate how relative measurements between fixed wing unmanned aerial vehicles (UAVs) and known landmarks can be used to cooperatively localize UAVs when GPS signals are not available. A centralized Extended Kalman Filter (EKF) is combine local sensor information from all the estimate required states of UAVs. We compare localization accuracy cooperative algorithm for different such as range, bearing, range rate, line-of-sight rate. The dynamics, IMU, airspeed, altimeter...
Recent advances in the fields of robotics and automation have spurred significant interest robust state estimation. To enable estimation, several methodologies been proposed. One such technique, which has shown promising performance, is concept iteratively estimating a Gaussian Mixture Model (GMM), based upon estimation residuals, to characterize measurement uncertainty model. Through this iterative process, model more accurately characterized, enables through appropriate de-weighting...
<h3>Abstract</h3> One of the fundamental problems robotics and navigation is estimation relative pose an external object with respect to observer. A common method for computing iterative closest point (ICP) algorithm, where a reference cloud known registered against sensed determine pose. To use this computed information in downstream processing algorithms, it necessary estimate uncertainty ICP output, typically represented as covariance matrix. In paper, novel estimating from data introduced.
This paper presents an innovative method for estimating wind velocity using optical ∞ow sensor mounted on a miniature air vehicle (MAV). Using the of features measured by in longitudinal and lateral directions, MAV’s crab-angle is estimated. By combining with measurements ground track from GPS airspeed, computed. Our algorithm allow this computation to be performed real time board MAV. Unlike previous techniques, approach does not require use magnetometers. has been implemented its...