检索结果-内蒙古大学图书馆

Pose-Graph Attentional Graph Neural Network for Lidar Place Recognition

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ramezani, Milad Wang, Liang Knights, Joshua Li, Zhibin Pounds, Pauline Moghadam, Peyman Robotics and Autonomous Systems DATA61 CSIRO Australia The University of Queensland Brisbane Australia Brisbane Australia

This paper proposes a pose-graph attentional graph neural network, called P-GAT, which compares (key)nodes between sequential and non-sequential sub-graphs for place recognition tasks as opposed to a common frame-to-frame retrieval problem formulation currently implemented in SOTA place recognition methods. P-GAT uses the maximum spatial and temporal information between neighbour cloud descriptors — generated by an existing encoder— utilising the concept of pose-graph SLAM. Leveraging intra- and inter-attention and graph neural network, P-GAT relates point clouds captured in nearby locations in Euclidean space and their embeddings in feature space. Experimental results on the large-scale publically available datasets demonstrate the effectiveness of our approach in scenes lacking distinct features and when training and testing environments have different distributions (domain adaptation). Further, an exhaustive comparison with the state-of-the-art shows improvements in performance gains. Code is available at https://***/csiro-robotics/P-GAT. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

Spectral Geometric Verification: Re-Ranking Point Cloud Retrieval for Metric Localization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Vidanapathirana, Kavisha Moghadam, Peyman Sridharan, Sridha Fookes, Clinton Robotics and Autonomous Systems Group DATA61 Csiro Australia Brisbane Australia

In large-scale metric localization, an incorrect result during retrieval will lead to an incorrect pose estimate or loop closure. Re-ranking methods propose to take into account all the top retrieval candidates and re-order them to increase the likelihood of the top candidate being correct. However, state-of-the-art re-ranking methods are inefficient when re-ranking many potential candidates due to their need for resource intensive point cloud registration between the query and each candidate. In this work, we propose an efficient spectral method for geometric verification (named SpectralGV) that does not require registration. We demonstrate how the optimal intercluster score of the correspondence compatibility graph of two point clouds represents a robust fitness score measuring their spatial consistency. This score takes into account the subtle geometric differences between structurally similar point clouds and therefore can be used to identify the correct candidate among potential matches retrieved by global similarity search. SpectralGV is deterministic, robust to outlier correspondences, and can be computed in parallel for all potential candidates. We conduct extensive experiments on 5 large-scale datasets to demonstrate that SpectralGV outperforms other state-of-the-art re-ranking methods and show that it consistently improves the recall and pose estimation of 3 state-of-the-art metric localization architectures while having a negligible effect on their runtime. The open-source implementation and trained models are available at: https://***/csiro-robotics/SpectralGV. Copyright © 2022, The Authors. All rights reserved.

关键词： Large dataset

autonomous Active Mapping in Steep Alpine Environments with Fixed-wing Aerial Vehicles

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lim, Jaeyoung Achermann, Florian Lawrance, Nicholas Siegwart, Roland Autonomous Systems Lab ETH Zürich Zürich8092 Switzerland CSIRO Robotics Data61 QLD4069 Australia

Monitoring large scale environments is a crucial task for managing remote alpine environments, especially for hazardous events such as avalanches. One key information for avalanche risk forecast is imagery of released avalanches. As these happen in remote and potentially dangerous locations this data is difficult to obtain. Fixed-wing vehicles, due to their long range and travel speeds are a promising platform to gather aerial imagery to map avalanche activities. However, operating such vehicles in mountainous terrain remains a challenge due to the complex topography, regulations, and uncertain environment. In this work, we present a system that is capable of safely navigating and mapping an avalanche using a fixed-wing aerial system and discuss the challenges arising when executing such a mission. We show in our field experiments that we can effectively navigate in steep terrain environments while maximizing the map quality. We expect our work to enable more autonomous operations of fixed-wing vehicles in alpine environments to maximize the quality of the data gathered. Copyright © 2024, The Authors. All rights reserved.

关键词： Fixed wings

Wildcat: Online Continuous-Time 3D Lidar-Inertial SLAM

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ramezani, Milad Khosoussi, Kasra Catt, Gavin Moghadam, Peyman Williams, Jason Borges, Paulo Pauling, Fred Kottege, Navinda The Robotics and Autonomous Systems Group DATA61 Csiro BrisbaneQLD4069 Australia

We present Wildcat, a novel online 3D lidar-inertial SLAM system with exceptional versatility and robustness. At its core, Wildcat combines a robust real-time lidar-inertial odometry module, utilising a continuous-time trajectory representation, with an efficient pose-graph optimisation module that seamlessly supports both the single- and multi-agent settings. The robustness of Wildcat was recently demonstrated in the DARPA Subterranean Challenge where it outperformed other SLAM systems across various types of sensing-degraded and perceptually challenging environments. In this paper, we extensively evaluate Wildcat in a diverse set of new and publicly available real-world datasets and showcase its superior robustness and versatility over two existing state-of-the-art lidar-inertial SLAM systems. Copyright © 2022, The Authors. All rights reserved.

关键词： Continuous time systems

SPECTRAL-ENHANCED TRANSFORMERS: LEVERAGING LARGE-SCALE PRETRAINED MODELS FOR HYPERSPECTRAL OBJECT TRACKING

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Mohamed, Shaheer Fernando, Tharindu Sridharan, Sridha Moghadam, Peyman Fookes, Clinton Signal Processing Artificial Intelligence and Vision Technologies Queensland University of Technology Brisbane Australia Robotics and Autonomous Systems Data61 CSIRO BrisbaneQLD Australia

Hyperspectral object tracking using snapshot mosaic cameras is emerging as it provides enhanced spectral information alongside spatial data, contributing to a more comprehensive understanding of material properties. Using transformers, which have consistently outperformed convolutional neural networks (CNNs) in learning better feature representations, would be expected to be effective for Hyperspectral object tracking. However, training large transformers necessitates extensive datasets and prolonged training periods. This is particularly critical for complex tasks like object tracking, and the scarcity of large datasets in the hyperspectral domain acts as a bottleneck in achieving the full potential of powerful transformer models. This paper proposes an effective methodology that adapts large pretrained transformer-based foundation models for hyperspectral object tracking. We propose an adaptive, learnable spatial-spectral token fusion module that can be extended to any transformer-based backbone for learning inherent spatial-spectral features in hyperspectral data. Furthermore, our model incorporates a cross-modality training pipeline that facilitates effective learning across hyperspectral datasets collected with different sensor modalities. This enables the extraction of complementary knowledge from additional modalities, whether or not they are present during testing. Our proposed model also achieves superior performance with minimal training iterations. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

Residual Skill Policies: Learning an Adaptable Skill-based Action Space for Reinforcement Learning for robotics

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Rana, Krishan Xu, Ming Tidd, Brendan Milford, Michael Sünderhauf, Niko QUT Centre for Robotics Queensland University of Technology Australia Data61 Robotics and Autonomous Systems Group CSIRO

Skill-based reinforcement learning (RL) has emerged as a promising strategy to leverage prior knowledge for accelerated robot learning. Skills are typically extracted from expert demonstrations and are embedded into a latent space from which they can be sampled as actions by a high-level RL agent. However, this skill space is expansive, and not all skills are relevant for a given robot state, making exploration difficult. Furthermore, the downstream RL agent is limited to learning structurally similar tasks to those used to construct the skill space. We firstly propose accelerating exploration in the skill space using state-conditioned generative models to directly bias the high-level agent towards only sampling skills relevant to a given state based on prior experience. Next, we propose a low-level residual policy for fine-grained skill adaptation enabling downstream RL agents to adapt to unseen task variations. Finally, we validate our approach across four challenging manipulation tasks that differ from those used to build the skill space, demonstrating our ability to learn across task variations while significantly accelerating exploration, outperforming prior works. © 2022, CC BY.

关键词： Reinforcement learning

Fisher Information Based Active Planning for Aerial Photogrammetry

学校读者我要写书评

暂无评论

Fisher Information Based Active Planning for Aerial Photogra...

IEEE International Conference on robotics and Automation (ICRA)

作者： Jaeyoung Lim Nicholas Lawrance Florian Achermann Thomas Stastny Rik Bähnemann Roland Siegwart Autonomous Systems Lab ETH Zürich Zürich Switzerland The Robotics and Autonomous Systems Group CSIRO Data61 QLD Australia Auterion AG Zürich Switzerland

Small uncrewed aerial systems (sUASs) are useful tools for 3D reconstruction due to their speed, ease of use, and ability to access high-utility viewpoints. Today, most aerial survey approaches generate a preplanned coverage pattern assuming a planar target region. However, this is inefficient since it results in superfluous overlap and suboptimal viewing angles and does not utilize the entire flight envelope. In this work, we propose active path planning for photogrammetric reconstruction. Our main contribution is a view utility function based on Fisher information approximating the offline reconstruction uncertainty. The metric enables online path planning to make in-flight decisions to collect geometrically informative image data in complex terrain. We evaluate our approach in a photorealistic simulation. A viewpoint selection study shows that our metric leads to faster and more precise reconstruction than state-of-the-art active planning metrics and adapts to different camera resolutions. Comparing our online planning approach to an ordinary fixed-wing aerial survey yields 3.2 × faster coverage of 16 ha undulated terrain without sacrificing precision.

关键词：

Deep Robust Multi-Robot Re-Localisation in Natural Environments

学校读者我要写书评

暂无评论

Deep Robust Multi-Robot Re-Localisation in Natural Environme...

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Milad Ramezani Ethan Griffiths Maryam Haghighat Alex Pitt Peyman Moghadam Robotics and Autonomous Systems DATA61 CSIRO Brisbane QLD Australia Queensland University of Technology (QUT) Brisbane Australia

The success of re-localisation has crucial implications for the practical deployment of robots operating within a prior map or relative to one another in real-world scenarios. Using single-modality, place recognition and localisation can be compromised in challenging environments such as forests. To address this, we propose a strategy to prevent lidar-based re-localisation failure using lidar-image cross-modality. Our solution relies on self-supervised 2D-3D feature matching to predict alignment and misalignment. Leveraging a deep network for lidar feature extraction and relative pose estimation between point clouds, we train a model to evaluate the estimated transformation. A model predicting the presence of misalignment is learned by analysing image-lidar similarity in the embedding space and the geometric constraints available within the region seen in both modalities in Euclidean space. Experimental results using real datasets (offline and online modes) demonstrate the effectiveness of the proposed pipeline for robust re-localisation in unstructured, natural environments.

关键词：

Spectral-Enhanced Transformers: Leveraging Large-Scale Pretrained Models for Hyperspectral Object Tracking

学校读者我要写书评

暂无评论

Spectral-Enhanced Transformers: Leveraging Large-Scale Pretr...

Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, WHISPERS

作者： Shaheer Mohamed Tharindu Fernando Sridha Sridharan Peyman Moghadam Clinton Fookes Signal Processing Artificial Intelligence and Vision Technologies Queensland University of Technology Brisbane Australia Robotics and Autonomous Systems Data61 CSIRO Brisbane QLD Australia

ISBN: (数字)9798331513139

ISBN: (纸本)9798331513146

关键词： Training Adaptation models Pipelines Transformers Feature extraction data models Spatial databases Object tracking Hyperspectral imaging Testing