Search

This document presents a framework for lidar-inertial localisation and mapping named 2Fast-2Lamaa. The method revolves around two main steps which are the inertial-aided undistortion of the lidar data and the scan-to-map registration using a distance-field representation of the environment. The initialisation-free undistortion uses inertial data to constrain the continuous trajectory of the sensor during the lidar scan. The eleven DoFs that fully characterise the trajectory are estimated by minimising lidar point-to-line and point-to-plane distances in a non-linear least-square formulation. The registration uses a map that provides a distance field for the environment based on Gaussian Process regression. The pose of an undistorted lidar scan is optimised to minimise the distance field queries of its points with respect to the map. After registration, the new geometric information is efficiently integrated into the map. The soundness of 2Fast-2Lamaa is demonstrated over several datasets (qualitative evaluation only). The real-time implementation is made publicly available at https://github.com/UTS-RI/2fast2lamaa.

Over the past decade, lidars have become a cornerstone of robotics state estimation and perception thanks to their ability to provide accurate geometric information about their surroundings in the form of 3D scans. Unfortunately, most of nowadays lidars do not take snapshots of the environment but sweep the environment over a period of time (typically around 100 ms). Such a rolling-shutter-like mechanism introduces motion distortion into the collected lidar scan, thus hindering downstream perception applications. In this paper, we present a novel method for motion distortion correction of lidar data by tightly coupling lidar with Inertial Measurement Unit (IMU) data. The motivation of this work is a map-free dynamic object detection based on lidar. The proposed lidar data undistortion method relies on continuous preintegrated of IMU measurements that allow parameterising the sensors' continuous 6-DoF trajectory using solely eleven discrete state variables (biases, initial velocity, and gravity direction). The undistortion consists of feature-based distance minimisation of point-to-line and point-to-plane residuals in a non-linear least-square formulation. Given undistorted geometric data over a short temporal window, the proposed pipeline computes the spatiotemporal normal vector of each of the lidar points. The temporal component of the normals is a proxy for the corresponding point's velocity, therefore allowing for learning-free dynamic object classification without the need for registration in a global reference frame. We demonstrate the soundness of the proposed method and its different components using public datasets and compare them with state-of-the-art lidar-inertial state estimation and dynamic object detection algorithms., Comment: Paper presented at the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

This paper presents DynORecon, a Dynamic Object Reconstruction system that leverages the information provided by Dynamic SLAM to simultaneously generate a volumetric map of observed moving entities while estimating free space to support navigation. By capitalising on the motion estimations provided by Dynamic SLAM, DynORecon continuously refines the representation of dynamic objects to eliminate residual artefacts from past observations and incrementally reconstructs each object, seamlessly integrating new observations to capture previously unseen structures. Our system is highly efficient (~20 FPS) and produces accurate (~10 cm) reconstructions of dynamic objects using simulated and real-world outdoor datasets., Comment: 7 pages, 6 figures, submitted to ICRA 2025

In-orbit automated servicing is a promising path towards lowering the cost of satellite operations and reducing the amount of orbital debris. For this purpose, we present a pipeline for automated satellite docking port detection and state estimation using monocular vision data from standard RGB sensing or an event camera. Rather than taking snapshots of the environment, an event camera has independent pixels that asynchronously respond to light changes, offering advantages such as high dynamic range, low power consumption and latency, etc. This work focuses on satellite-agnostic operations (only a geometric knowledge of the actual port is required) using the recently released Lockheed Martin Mission Augmentation Port (LM-MAP) as the target. By leveraging shallow data-driven techniques to preprocess the incoming data to highlight the LM-MAP's reflective navigational aids and then using basic geometric models for state estimation, we present a lightweight and data-efficient pipeline that can be used independently with either RGB or event cameras. We demonstrate the soundness of the pipeline and perform a quantitative comparison of the two modalities based on data collected with a photometrically accurate test bench that includes a robotic arm to simulate the target satellite's uncontrolled motion., Comment: Submitted to IEEE ICRA 2025

We consider the problem of planning collision-free trajectories on distance fields. Our key observation is that querying a distance field at one configuration reveals a region of safe space whose radius is given by the distance value, obviating the need for additional collision checking within the safe region. We refer to such regions as safe bubbles, and show that safe bubbles can be obtained from any Lipschitz-continuous safety constraint. Inspired by sampling-based planning algorithms, we present three algorithms for constructing a safe bubble cover of free space, named bubble roadmap (BRM), rapidly exploring bubble graph (RBG), and expansive bubble graph (EBG). The bubble sampling algorithms are combined with a hierarchical planning method that first computes a discrete path of bubbles, followed by a continuous path within the bubbles computed via convex optimization. Experimental results show that the bubble-based methods yield up to 5- 10 times cost reduction relative to conventional baselines while simultaneously reducing computational efforts by orders of magnitude., Comment: 16 pages, 11 figures. Submitted to International Symposium on Robotics Research 2024

Robots reason about the environment through dedicated representations. Popular choices for dense representations exploit Truncated Signed Distance Functions (TSDF) and Octree data structures. However, TSDF is a projective signed distance obtained directly from depth measurements that overestimates the Euclidean distance. Octrees, despite being memory efficient, require tree traversal and can lead to increased runtime in large scenarios. Other representations based on Gaussian Process (GP) distance fields are appealing due to their probabilistic and continuous nature, but the computational complexity is a concern. In this paper, we present an online efficient mapping framework that seamlessly couples GP distance fields and the fast-access VDB data structure. This framework incrementally builds the Euclidean distance field and fuses other surface properties, like intensity or colour, into a global scene representation that can cater for large-scale scenarios. The key aspect is a latent Local GP Signed Distance Field (L-GPDF) contained in a local VDB structure that allows fast queries of the Euclidean distance, surface properties and their uncertainties for arbitrary points in the field of view. Probabilistic fusion is then performed by merging the inferred values of these points into a global VDB structure that is efficiently maintained over time. After fusion, the surface mesh is recovered, and a global GP Signed Distance Field (G-GPDF) is generated and made available for downstream applications to query accurate distance and gradients. A comparison with the state-of-the-art frameworks shows superior efficiency and accuracy of the inferred distance field and comparable reconstruction performance. The accompanying code will be publicly available. https://github.com/UTS-RI/VDB_GPDF

For the three body problem with equal masses, we prove that the most symmetric continuation class of Lagrange's equilateral triangle solution, also referred to as the $P_{12}$ family of Marchal, contains the remarkable figure eight choreography discovered by Moore in 1993, and proven to exist by Chenciner and Montgomery in 2000. This settles a conjecture of Marchal which dates back to the 1999 conference on Celestial Mechanics in Evanston Illinois, celebrating Donald Saari's 60th birthday.

Easy Read text is one of the main forms of access to information for people with reading difficulties. One of the key characteristics of this type of text is the requirement to split sentences into smaller grammatical segments, to facilitate reading. Automated segmentation methods could foster the creation of Easy Read content, but their viability has yet to be addressed. In this work, we study novel methods for the task, leveraging masked and generative language models, along with constituent parsing. We conduct comprehensive automatic and human evaluations in three languages, analysing the strengths and weaknesses of the proposed alternatives, under scarce resource limitations. Our results highlight the viability of automated Easy Read text segmentation and remaining deficiencies compared to expert-driven human segmentation.

The prominent role of electron-electron interactions in two-dimensional (2D) materials versus three-dimensional (3D) ones is at the origin of the great variety of fermionic correlated states reported in the literature. In this respect, artificial van der Waals heterostructures comprising single layers of highly correlated insulators allow one to explore the effect of the subtle interlayer interaction in the way electrons correlate. In this work, we study the temperature dependence of the electronic properties of a van der Waals heterostructure composed of a single-layer Mott insulator lying on a metallic substrate by performing quasi-particle interference (QPI) maps. We show the emergence of a Fermi contour in the 2D Mott insulator at temperatures below 11K, which we attribute to the delocalization of the Mott electrons associated with the formation of a quantum coherent Kondo lattice. This Kondo lattice introduces a new periodicity in the system, so that the resulting Fermi surface encompasses both the substrate conduction electrons and the now delocalized correlated electrons from the 2D Mott insulator. Density Functional Theory calculations allow us to pinpoint the scattering vectors responsible for the experimentally observed quasi-particle interference maps, thus providing a complete picture of the delocalization of highly correlated electrons in a 2D Mott insulator., Comment: 15 pages, 4 figures

Human-robot collaborative applications require scene representations that are kept up-to-date and facilitate safe motions in dynamic scenes. In this letter, we present an interactive distance field mapping and planning (IDMP) framework that handles dynamic objects and collision avoidance through an efficient representation. We define interactive mapping and planning as the process of creating and updating the representation of the scene online while simultaneously planning and adapting the robot's actions based on that representation. The key aspect of this work is an efficient Gaussian Process field that performs incremental updates and handles dynamic objects reliably by identifying moving points via a simple and elegant formulation based on queries from a temporary latent model. In terms of mapping, IDMP is able to fuse point cloud data from single and multiple sensors, query the free space at any spatial resolution, and deal with moving objects without semantics. In terms of planning, IDMP allows seamless integration with gradient-based reactive planners facilitating dynamic obstacle avoidance for safe human-robot interactions. Our mapping performance is evaluated on both real and synthetic datasets. A comparison with similar state-of-the-art frameworks shows superior performance when handling dynamic objects and comparable or better performance in the accuracy of the computed distance and gradient field. Finally, we show how the framework can be used for fast motion planning in the presence of moving objects both in simulated and real-world scenes. An accompanying video, code, and datasets are made publicly available https://uts-ri.github.io/IDMP.

In this paper, we introduce the configuration space of orbits, a generalization of the configuration space of points but for algebraic varieties that are acted by an algebraic reductive group. We develop a novel method for computing the $S_n$-equivariant $E$-polynomial of an algebraic variety, and we apply it to this new kind of varieties., Comment: 22 pages. Comments are welcome

This study delves into the intriguing properties of 1H/1T-TaS$_2$ van der Waals heterostructure, focusing on the transparency of the 1H layer to the Charge Density Wave of the underlying 1T layer. Despite the sizable interlayer separation and metallic nature of the 1H layer, positive bias voltages result in a pronounced superposition of the 1T charge density wave structure on the 1H layer. The conventional explanation relying on tunneling effects proves insufficient. Through a comprehensive investigation combining lowtemperature scanning tunneling microscopy, scanning tunneling spectroscopy, non-contact atomic force microscopy, and firstprinciples calculations, we propose an alternative interpretation. The transparency effect arises from a weak yet substantial electronic coupling between the 1H and 1T layers, challenging prior understanding of the system. Our results highlight the critical role played by interlayer electronic interactions in van der Waals heterostructures to determine the final ground states of the systems.

This work presents an experimental and theoretical insight into formation mechanisms of single crystalline wurtzite InN quantum dots (QDs) fabricated via metal droplet epitaxy (DE) by employing plasma assisted molecular beam epitaxy. The applied procedure consists of two fabrication stages. During the first stage, the cold substrate (T = 15 {\deg}C) is exposed to an impinging In flux, resulting in formation of metallic In droplets on the substrate surface, and then to an impinging active nitrogen flux, resulting in In conversion into polycrystalline InN islands. During the second stage, the substrate, which is still kept exposed to active nitrogen, is heated up to T = 300 {\deg}C, to allow for the reorganization of extended polycrystalline InN islands into groups of independent single crystalline wurtzite InN QDs. This work provides a detailed experimental insight into both fabrication stages and their qualitative explanations within the scopes of adatom surface kinetics (stage I) and total energy per unit crystal volume minimization (stage II). Finally, the formation mechanisms of InN QDs on the three different substrates (Si(111), Si(001) and In0.3Ga0.7N/Si(111)) are compared, and also linked to the formation mechanisms of other more studied nanostructures, such as self assembled GaN/AlN QDs and self assembled and selective area grown GaN nanowires., Comment: 20 pages, 11 figures

InN quantum dots (QDs) are considered to be promising nanostructures for different device applications. For any hexagonal AB stacking semiconductor system, polarity is an important feature which affects the electronic properties. Therefore, the determination of this characteristic on any wurtzite (semi)polar III nitride compound or alloy is essential for defining its applicability. In this paper, the polarity of InN QDs grown on silicon by indium droplet epitaxy plus nitridation and annealing was determined by a modified approach combining exit wave reconstruction with negative spherical aberration high resolution lattice imaging using TEM. Comparing the micrographs of two QDs from the same TEM specimen with the simulated images of InN slab structures generated under the same conditions as of the experiments, it was confirmed that the QDs of the present study are N polar. Given that the settlement of material's polarity has always been a tedious, indirect and controversial issue, the major value of our proposal is to provide a straightforward procedure to determine the polar direction from atomic-resolution focal series images., Comment: 7 pages 5 figures

We investigate the axial and radial growth of GaN nanowires upon a variation of the Ga flux during molecular beam epitaxial growth. An increase in the Ga flux promotes radial growth without affecting the axial growth rate. In contrast, a decrease in the Ga flux reduces the axial growth rate without any change in the radius. These results are explained by a kinetic growth model that accounts for both the diffusion of Ga adatoms along the side facets towards the nanowire tip and the finite amount of active N available for the growth. The model explains the formation of a new equilibrium nanowire radius after increasing the Ga flux and provides an explanation for two well known but so far not understood experimental facts: the necessity of effectively N-rich conditions for the spontaneous growth of GaN nanowires and the increase in nanowire radius with increasing III/V flux ratios.

Indium incorporation and surface morphology of InAlN layers grown on (0001)GaN by plasma-assisted molecular beam epitaxy were investigated as a function of the impinging In flux and the substrate temperature in the 450-610$^{\circ}$C range. In incorporation was found to decrease with substrate temperature due to thermal decomposition of the growing layer, while for a given temperature it increased with the impinging In flux until stoichiometry was reached at the growth front. The InN losses during growth followed an Arrhenius behaviour characterized by an activation energy of 2.0 eV. A growth diagram highly instrumental to identify optimum growth conditions was established.

Real-time analysis of the growth modes during homoepitaxial (0001)GaN growth by plasma-assisted molecular beam epitaxy was performed using reflection high energy electron diffraction. A growth mode map was established as a function of Ga/N flux ratio and growth temperature, exhibiting distinct transitions between three-dimensional (3D), layer-by-layer and step-flow growth mode. The layer-by-layer to step-flow growth transition under Ga-rich growth was surfactant mediated and related to a Ga adlayer coverage of one monolayer. Under N-rich conditions the transition from 3D to layer-by-layer growth was predominantly thermally activated, facilitating two-dimensional growth at temperatures of thermal decomposition.

Thermal decomposition of wurtzite (0001)-oriented GaN was analyzed: in vacuum, under active N exposure, and during growth by rf-plasma assisted molecular beam epitaxy. The GaN decomposition rate was determined by measurements of the Ga desorption using in situ quadrupole mass spectrometry, which showed Arrhenius behavior with an apparent activation energy of 3.1 eV. Clear signatures of intensity oscillations during reflection high-energy electron diffraction measurements facilitated complementary evaluation of the decomposition rate and highlighted a layer-by-layer decomposition mode in vacuum. Exposure to active nitrogen, either under vacuum or during growth under N-rich growth conditions, strongly reduced the GaN losses due to GaN decomposition.

Room temperature photoluminescence and optical absorption spectra have been measured in wurtzite In$_{x}$Al$_{y}$Ga$_{1-x-y}$N (x $\approx$ 0.06, 0.02 < y < 0.27) layers grown by molecular beam epitaxy. Photoluminescence spectra show both an enhancement of the integrated intensity and an increasing Stokes shift with the Al content. Both effects arise from an Al-enhanced exciton localization revealed by the S- and W-shaped temperature dependences of the photoluminescence emission energy and bandwidth respectively. Present results point to these materials as a promising choice for the active region in efficient light emitters. An In-related bowing parameter of 1.6 eV was derived from optical absorption data.

Indium incorporation into wurtzite (0001)-oriented In$_{x}$Al$_{y}$Ga$_{1-x-y}$N layers grown by plasma-assisted molecular beam epitaxy was studied as a function of the growth temperature (565-635 $^{\circ}$C) and the AlN mole fraction (0.01 < y < 0.27). The layer stoichiometry was determined by Rutherford backscattering spectrometry (RBS). RBS shows that Indium incorporation decreased continuously with increasing growth temperature due to thermally enhanced dissociation of In-N bonds and for increasing AlN mole fractions. High resolution x-ray diffraction and transmission electron microscopy (TEM) measurements did not show evidence of phase separation. The mosaicity of the quaternary layers was found to be mainly determined by the growth temperature and independent on alloy composition within the range studied. However, depending on the AlN mole fraction, nanometer-sized composition fluctuations were detected by TEM. Photoluminescence spectra showed a single broad emission at room temperature, with energy and bandwidth S- and W-shaped temperature dependences typical of exciton localization by alloy inhomogeneities. Cathodo-luminescence measurements demonstrated that the alloy inhomogeneities, responsible of exciton localization, occur on a lateral length scale below 150 nm, which is corroborated by TEM.

The morphology of GaN samples grown by plasma-assisted molecular beam epitaxy on Si(111) was systematically studied as a function of impinging Ga/N flux ratio and growth temperature (750-850{\deg}C).Two different growth regimes were identified: compact and nanocolumnar. A growth diagram was established as a function of growth parameters, exhibiting the transition between growth regimes, and showing under which growth conditions GaN cannot be grown due to thermal decomposition and Ga desorption. Present results indicate that adatoms diffusion length and the actual Ga/N ratio on the growing surface are key factors to achieve nanocolumnar growth., Comment: This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in (citation of published article) and may be found at J. Appl. Phys. 106, 126102 (2009) and may be found at https://doi.org/10.1063/1.3267151

The Split and Rephrase (SPRP) task, which consists in splitting complex sentences into a sequence of shorter grammatical sentences, while preserving the original meaning, can facilitate the processing of complex texts for humans and machines alike. It is also a valuable testbed to evaluate natural language processing models, as it requires modelling complex grammatical aspects. In this work, we evaluate large language models on the task, showing that they can provide large improvements over the state of the art on the main metrics, although still lagging in terms of splitting compliance. Results from two human evaluations further support the conclusions drawn from automated metric results. We provide a comprehensive study that includes prompting variants, domain shift, fine-tuned pretrained language models of varying parameter size and training data volumes, contrasted with both zero-shot and few-shot approaches on instruction-tuned language models. Although the latter were markedly outperformed by fine-tuned models, they may constitute a reasonable off-the-shelf alternative. Our results provide a fine-grained analysis of the potential and limitations of large language models for SPRP, with significant improvements achievable using relatively small amounts of training data and model parameters overall, and remaining limitations for all models on the task.

Reference-guided DNA sequencing and alignment is an important process in computational molecular biology. The amount of DNA data grows very fast, and many new genomes are waiting to be sequenced while millions of private genomes need to be re-sequenced. Each human genome has 3.2 B base pairs, and each one could be stored with 2 bits of information, so one human genome would take 6.4 B bits or about 760 MB of storage (National Institute of General Medical Sciences). Today most powerful tensor processing units cannot handle the volume of DNA data necessitating a major leap in computing power. It is, therefore, important to investigate the usefulness of quantum computers in genomic data analysis, especially in DNA sequence alignment. Quantum computers are expected to be involved in DNA sequencing, initially as parts of classical systems, acting as quantum accelerators. The number of available qubits is increasing annually, and future quantum computers could conduct DNA sequencing, taking the place of classical computing systems. We present a novel quantum algorithm for reference-guided DNA sequence alignment modeled with gate-based quantum computing. The algorithm is scalable, can be integrated into existing classical DNA sequencing systems and is intentionally structured to limit computational errors. The quantum algorithm has been tested using the quantum processing units and simulators provided by IBM Quantum, and its correctness has been confirmed., Comment: 19 pages, 13 figures

De novo DNA sequence assembly is based on finding paths in overlap graphs, which is a NP-hard problem. We developed a quantum algorithm for de novo assembly based on quantum walks in graphs. The overlap graph is partitioned repeatedly to smaller graphs that form a hierarchical structure. We use quantum walks to find paths in low rank graphs and a quantum algorithm that finds Hamiltonian paths in high hierarchical rank. We tested the partitioning quantum algorithm, as well as the quantum algorithm that finds Hamiltonian paths in high hierarchical rank and confirmed its correct operation using Qiskit. We developed a custom simulation for quantum walks to search for paths in low rank graphs. The approach described in this paper may serve as a basis for the development of efficient quantum algorithms that solve the de novo DNA assembly problem., Comment: 13 pages, 7 figures