A5103146107- Target Tracking and Data Fusion in Sensor Networks
- Music Technology and Sound Studies
- Teaching and Learning Programming
- Gaussian Processes and Bayesian Inference
- Infrared Target Detection Methodologies
- Robotics and Sensor-Based Localization
- Music and Audio Processing
- Indoor and Outdoor Localization Technologies
- Advanced Multi-Objective Optimization Algorithms
- Musicology and Musical Analysis
- Modular Robots and Swarm Intelligence
- Embedded Systems Design Techniques
- Hydraulic and Pneumatic Systems
- Speech and Audio Processing
- Smart Agriculture and AI
- Software Testing and Debugging Techniques
- Diverse Musicological Studies
- Probabilistic and Robust Engineering Design
- Robot Manipulation and Learning
- Computer Graphics and Visualization Techniques
- Software Engineering Research
- Distributed Sensor Networks and Detection Algorithms
- Neuroscience and Music Perception
- Guidance and Control Systems
- Advanced Malware Detection Techniques
Queensland University of Technology
2017-2021
Microsoft (United States)
2014
Qinetiq (United Kingdom)
2007
Virginia Tech
2007
Xerox (France)
2001
Johns Hopkins University
1999-2000
LIG Science (United States)
1998-1999
Robotic challenges like the Amazon Picking Challenge (APC) or DARPA Challenges are an established and important way to drive scientific progress. They make research comparable on a well-defined benchmark with equal test conditions for all participants. However, such challenge events occur only occasionally, limited small number of contestants, very difficult replicate after main event. We present new physical robotic picking: ACRV Benchmark. Designed be reproducible, it consists set 42...
We present our findings on the state of field algorithm visualization, based extensive search and analysis links to hundreds visualizations. seek answer questions such as how content is distributed among topics, who created visualizations when, overall quality available visualizations, are disseminated. have built a wiki that currently catalogs over 350 contains beginnings an annotated bibliography visualization literature, provides information about researchers projects. Unfortunately, we...
Algorithm visualizations are widely viewed as having the potential for major impact on computer science education, but their quality is highly variable. We report software development practices used by creators of algorithm visualizations, based data that can be inferred from a catalog over 600 visualizations. Since nearly all free use and many provide source code, they might construed being open software. Yet AV developers do not appear to have best practices. discuss how such employed...
In our earlier work, we focused on pose estimation of ground- based targets as viewed via forward-looking passive infrared (FLIR) systems and laser radar (LADAR) imaging sensors. this paper, will study individual joint sensor performance to provide a more complete understanding suite. We also the addition high range- resolution (HRR). Data from these three sensors are simulated using CAD models for interest in conjunction with XPATCH range simulation software, Silicon Graphics workstations...
This paper explores the parameter estimation problem for a differential-drive agricultural vehicle with unknown parameters. The configuration is commonly used in robotic applications as well large machinery such harvesters. We use simulation scenarios to compare performance of two dual filters and state estimation: Dual Liu West filter (D-L&WF) novel Merging particle (D-MPF). Our initial results indicate slightly better D-MPF we discuss limitations advantages each filter. Dual-particle...
Our work focuses on pose estimation of ground-based targets viewed via multiple sensors including forward-looking infrared radar (FLIR) systems and laser (LADAR) range imagers. Data from these two are simulated using CAD models for the interest in conjunction with Silicon Graphics workstations, PRISM simulation package, statistical model LADAR described by Green Shapiro. Using a Bayesian framework, we quantitatively examine both pose-dependent variations performance, relative performance...
We have been studying information theoretic measures, entropy and mutual information, as performance bounds on the gain given a standard suite of sensors. Object pose is described by single angle rotation using Lie group parameterization; observations are simulated CAD models for targets interest simulators such PRISM infrared simulator. Variability in data due to sensor which scene remotely observed statistically characterized via likelihood function. Taking Bayesian approach, inference...
The Multiple Airborne Sensor Targeting and Evaluation Rig (MASTER) is a high fidelity simulation environment in which data fusion, tracking sensor management algorithms developed within QinetiQ Ltd. can be demonstrated evaluated. In this paper we report an observer trajectory planning tool that adds considerable functionality to MASTER. This coordinate multiple platforms highly manoeuvring targets. It does by applying instantaneous thrusts each platform, the magnitude of chosen gain maximum...
Automated target recognition (ATR) is a problem of great importance in wide variety applications: from military to recognizing flow-patterns fluid- dynamics anatomical shape-studies. The basic goal utilize observations (images, signals) remote sensors (such as videos, radars, MRI or PET) identify the objects being observed. In statistical framework, probability distributions on parameters representing object unknowns are derived an analyzed compute inferences (please refer [1] for detailed...
We have been studying information theoretic measures, entropy and mutual information, as performance metrics for object recognition given a standard suite of sensors. Our work has focused on analysis the pose estimation ground-based objects viewed remotely via sensor suite. Target is described by single angle rotation using Lie group parameterization: O (epsilon) SO(2), 2 X matrices. Variability in data due to which scene observed statistically characterized likelihood function. Taking...
In this article, we propose two novel experimental design techniques for designing maximally informative experiments to estimate the parameters of nonlinear dynamical vehicle models. The include a batch and sequential technique that seek maximize expected Shannon information gain parameter distribution using either an online or offline approach (respectively). We apply compare in both simulation real-world with wheeled vehicle. our experiments, proposed designs provide superior gains...
The recognition of targets in infrared scenes is complicated by the wide variety appearances associated with different thermodynamic states. We represent variability signatures via an expansion terms 'eigentanks' derived from a principal component analysis performed over target's surface. Employing Poisson sensor likelihood, or equivalently likelihood based on Csiszar's I-divergence, natural discrepancy measure for nonnegative images, yields coupled set nonlinear equations which must be...