A5005183926- Robot Manipulation and Learning
- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
- Robotic Path Planning Algorithms
- Teleoperation and Haptic Systems
- Human-Automation Interaction and Safety
- Robotic Mechanisms and Dynamics
- Soft Robotics and Applications
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- Remote-Sensing Image Classification
- Robotics and Automated Systems
- Advancements in Battery Materials
- Infrared Target Detection Methodologies
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Experimental Learning in Engineering
- Advanced Battery Technologies Research
- Human Pose and Action Recognition
- Advanced Neural Network Applications
- Image Enhancement Techniques
- Advanced Multi-Objective Optimization Algorithms
- Hand Gesture Recognition Systems
- Reinforcement Learning in Robotics
- Modular Robots and Swarm Intelligence
- Artificial Immune Systems Applications
University of Birmingham
2016-2025
The Faraday Institution
2019-2025
Bristol Robotics Laboratory
2016-2024
Karlsruhe Institute of Technology
2023
Royal Society
2021
Stevens Institute of Technology
2005-2020
American Society For Engineering Education
2020
National Nuclear Laboratory
2018
Robotics Research (United States)
2018
Australian National University
2013
Visual tracking has attracted a significant attention in the last few decades. The recent surge number of publications on tracking-related problems have made it almost impossible to follow developments field. One reasons is that there lack commonly accepted annotated data-sets and standardized evaluation protocols would allow objective comparison different methods. To address this issue, Object Tracking (VOT) workshop was organized conjunction with ICCV2013. Researchers from academia as well...
Abstract The market dynamics, and their impact on a future circular economy for lithium-ion batteries (LIB), are presented in this roadmap, with safety as an integral consideration throughout the life cycle. At point of end-of-life (EOL), there is range potential options—remanufacturing, reuse recycling. Diagnostics play significant role evaluating state-of-health condition batteries, improvements to diagnostic techniques evaluated. present, manual disassembly dominates EOL disposal,...
This paper presents a method for one-shot learning of dexterous grasps and grasp generation novel objects. A model each type is learned from single kinesthetic demonstration several types are taught. These models used to select generate unfamiliar Both the stages use an incomplete point cloud depth camera, so no prior object shape used. The product experts, in which experts two types. first contact density over pose hand link relative local surface. second hand-configuration whole-hand...
This paper addresses the problem of finding sparse solutions to linear systems. Although this involves two competing cost function terms (measurement error and a sparsity-inducing term), previous approaches combine these into single term solve using conventional numerical optimization methods. In contrast, main contribution is use multiobjective approach. The begins by investigating reconstruction problem, presents data show that knee regions do exist on Pareto front (PF) for optimal can be...
Feature selection is an important approach for reducing the dimension of high-dimensional data. In recent years, many feature algorithms have been proposed, but most them only exploit information from data space. They often neglect useful contained in space, and do not make full use characteristics To overcome this problem, paper proposes a new unsupervised algorithm, called non-negative spectral learning sparse regression-based dual-graph regularized (NSSRD). NSSRD based on framework joint...
Deep learning has obtained state-of-the-art results in a variety of computer vision tasks and also been used to solve SAR image classification problems. algorithms typically require large amount training data achieve high accuracy. In contrast, the size datasets is often comparatively limited. Therefore, this paper proposes novel method, deep memory convolution neural networks (M-Net), alleviate problem overfitting caused by insufficient samples. Based on convolutional (CNN), M-Net adds an...
This paper addresses the problem of RGBD object recognition in real-world applications, where large amounts annotated training data are typically unavailable. To overcome this problem, we propose a novel, weakly-supervised learning architecture (DCNN-GPC) which combines parametric models (a pair Deep Convolutional Neural Networks (DCNN) for RGB and D modalities) with non-parametric (Gaussian Process Classification). Our system is initially trained using small amount labeled data, then...
Semantic segmentation of high-resolution remote sensing images is highly challenging due to the presence a complicated background, irregular target shapes, and similarities in appearance multiple categories. Most existing methods that rely only on simple fusion extracted multi-scale features often fail provide satisfactory results when there large difference sizes. Handling this problem through context extraction efficient features, paper we present an end-to-end adaptive feature network...
This paper addresses the problems of automatically planning autonomous underwater vehicle (AUV) paths which best exploit complex current data, from computational estuarine model forecasts, while also avoiding obstacles. In particular we examine possibilities for a novel type AUV mission deployment in fast flowing tidal river regions experience bi-directional flow. These environments are interesting that, by choosing an appropriate path space and time, may both bypass adverse currents too to...
We present early pilot-studies of a new international project, developing advanced robotics to handle nuclear waste. Despite enormous remote handling requirements, there has been remarkably little use robots by the industry. The few deployed have directly teleoperated in rudimentary ways, with no control methods or autonomy. Most is still done an aging workforce highly skilled experts, using 1960s style mechanical Master-Slave devices. In contrast, this paper explores how novice human...
This paper shows how a robot arm can follow and grasp moving objects tracked by vision system, as is needed when human hands over an object to the during collaborative working. While being arbitrarily moved co-worker, set of likely grasps, generated learned planner, are evaluated online generate feasible with respect both: current configuration respecting target grasp; constraints finding collision-free trajectory reach that configuration. A task-based cost function enables relaxation...
The fuzzy c-means (FCM) clustering algorithm has been widely used in image segmentation. However, FCM exhibits poor robustness to noise, often leading unsatisfactory segmentations on noisy images. Additionally, the is sensitive choice of initial cluster centers. In order solve these problems, this paper proposes clone kernel spatial (CKS_FCM), which improves segmentation performance several ways. First, CKS_FCM, an immune generate centers, helps prevent from converging local optima. Second,...
Semi-supervised non-negative matrix factorization (NMF) exploits the strengths of NMF in effectively learning local information contained data and is also able to achieve effective when only a small fraction labeled. particularly useful for dimensionality reduction high-dimensional data. However, mapping between low-dimensional representation, learned by semi-supervised NMF, original contains complex hierarchical structural information, which hard extract using single-layer clustering...
The capacitated arc routing problem (CARP) has attracted considerable attention from researchers due to its broad potential for social applications. This paper builds on, and develops beyond, the cooperative coevolutionary algorithm based on route distance grouping (RDG-MAENS), recently proposed by Mei et al. Although Mei's method proved superior previous algorithms, we discuss several remaining drawbacks propose solutions overcome them. First, although RDG is used in searching better...
This paper presents a novel robust method for single target tracking in RGB-D images, and also contributes substantial new benchmark dataset evaluating trackers. While object's color distribution is reasonably motion-invariant, this not true the target's depth distribution, which continually varies as moves relative to camera. It therefore nontrivial design models can fully exploit (potentially very rich) information tracking. For reason, much of previous literature relies on tracking, while...
Disassembly of electric vehicle batteries is a critical stage in recovery, recycling and re-use high-value battery materials, but complicated by limited standardisation, design complexity, compounded uncertainty safety issues from varying end-of-life condition. Telerobotics presents an avenue for semi-autonomous robotic disassembly that addresses these challenges. However, it suggested quality realism the user's haptic interactions with environment important precise, contact-rich...
An important problem in robotic manipulation is the ability to predict how objects behave under manipulative actions. This necessary allow planning of object manipulations. Physics simulators can be used do this, but they model many kinds interaction poorly. alternative learn a motion for by interacting with them. In this paper we address learning interactions rigid bodies probabilistic framework, and demonstrate results domain push manipulation. A robot arm applies random pushes various...
Trajectory optimization is an essential tool for motion planning under multiple constraints of robotic manipulators. Optimization-based methods can explicitly optimize a trajectory by leveraging prior knowledge the system and have been used in various applications such as collision avoidance. However, these often require hand-coded cost function order to achieve desired behavior. Specifying complex behavior, e.g., disentangling rope, nontrivial task that even infeasible. Learning from...