arXiv subjects

Hanwen Kang

Publications and source records attributed to Hanwen Kang.

At least 19 recordsLinked to original sources

arXiv2026

Aligning Heterogeneous DFT Datasets: A Graph Neural Network Approach to Cross-Functional Formation Energies

Heterogeneous density functional theory (DFT) calculations, particularly plane-wave implementations, introduce systematic formation energy errors ranging from tens to hundreds of meV/atom, depending on the selection of exchange-correlation functionals, kinetic energy cutoffs, pseudopotentials, and dispersion corrections. As demonstrated by the MatPES dataset, identical structures can exhibit an average energy discrepancy of 107 meV/atom between PBE and r2SCAN calculations. Such method-dependent discrepancies hinder the integration of multi-source DFT data, greatly limiting the scale and quality of datasets for training robust materials AI models. Here, we resolve this fundamental data silo barrier via graph-based transfer learning. Leveraging 380,190 structurally paired PBE-r2SCAN entries from the MatPES database, we train a structure-aware graph neural network to predict cross-functional energy residuals and align inconsistent DFT energy scales. By adopting GPTFF model architecture, the model converts conventional PBE energies to r2SCAN-level accuracy with a mean absolute error of 14.3 meV/atom, compared with 18.2 meV/atom achieved by CHGNet. This versatile approach effectively upgrades massive legacy PBE datasets to high-precision r2SCAN standards. It enables reliable predictions of phase stability, battery voltage profiles, and reaction thermodynamics, while allowing the integration of multi-source DFT data to advance the development of high-performance materials foundation models.

cond-mat.mtrl-sci↗

arXiv2026

Are Machine Learning Interatomic Potentials Truly Practical? A Benchmark of 23 Mainstream Models

Most MLIP benchmarks reward static accuracy while ignoring inference efficiency and hardware scalability -- driving model bloat with unclear real-world value. We benchmark 23 mainstream open-source MLIPs on a low-cost NVIDIA DGX Spark (128 GB native memory, capped at 80 GB to mimic ordinary lab hardware), using a fixed 192-atom system under a unified ASE-based pipeline. We evaluate three dimensions: predictive accuracy, MD simulation throughput, and atomic scalability. Our results expose a sharp accuracy-efficiency trade-off: large SOTA models deliver only 3-5 meV/atom more accuracy than lightweight ones, but lose orders of magnitude in throughput -- in the worst case, becoming only marginally faster than DFT itself. Lightweight MLIPs, by contrast, sit on the Pareto frontier and run on modest hardware. The lesson is that single-dimensional benchmarks mislead the field, and that future MLIP development should value efficiency and scalability alongside accuracy.

cond-mat.mtrl-sci↗

arXiv2026

RoBoSR: Structured Scene Representations for Embodied Robotic Reasoning

Despite rapid progress, embodied reasoning under real-world variability remains challenging. Existing approaches rely on demonstration-driven sequential biases, limiting flexibility in open-ended and long-horizon tasks that require structured reasoning over evolving states. We introduce RoBoSR, an intermediate structural representation that formulates manipulation as step-wise state transitions over semantically grounded, object-centric scene graphs. By modeling object states and their spatial relations at the perception-action interface, RoBoSR disentangles high-level task reasoning from raw inputs and enables structured reasoning over preconditions, effects, and goal states. This representation endows the agent with causal reasoning capability, enforcing subtask dependencies and supporting coherent long-horizon task planning. To learn such structure-aware reasoning, we construct Manip-Cognition-1.6M, an open-world dataset that jointly supervises scene understanding, instruction interpretation, and subtask planning across diverse tasks. Across several benchmarks and real-world demonstrations, our method consistently outperforms prompting-based methods and classical TAMP baselines in zero-shot generalization and long-horizon tasks. The results underscore structured intermediate representations as a critical inductive bias for scalable embodied reasoning.

cs.RO↗

arXiv2026

OMNI-PoseX: A Fast Vision Model for 6D Object Pose Estimation in Embodied Tasks

Accurate 6D object pose estimation is a fundamental capability for embodied agents, yet remains highly challenging in open-world environments. Many existing methods often rely on closed-set assumptions or geometry-agnostic regression schemes, limiting their generalization, stability, and real-time applicability in robotic systems. We present OMNI-PoseX, a vision foundation model that introduces a novel network architecture unifying open-vocabulary perception with an SO(3)-aware reflected flow matching pose predictor. The architecture decouples object-level understanding from geometry-consistent rotation inference, and employs a lightweight multi-modal fusion strategy that conditions rotation-sensitive geometric features on compact semantic embeddings, enabling efficient and stable 6D pose estimation. To enhance robustness and generalization, the model is trained on large-scale 6D pose datasets, leveraging broad object diversity, viewpoint variation, and scene complexity to build a scalable open-world pose backbone. Comprehensive evaluations across benchmark pose estimation, ablation studies, zero-shot generalization, and system-level robotic grasping integration demonstrate the effectiveness of OMNI-PoseX. The OMNI-PoseX achieves SOTA pose accuracy and real-time efficiency, while delivering geometrically consistent predictions that enable reliable grasping of diverse, previously unseen objects.

cs.RO↗

arXiv2026

GSR: Learning Structured Reasoning for Embodied Manipulation

Despite rapid progress, embodied agents still struggle with long-horizon manipulation that requires maintaining spatial consistency, causal dependencies, and goal constraints. A key limitation of existing approaches is that task reasoning is implicitly embedded in high-dimensional latent representations, making it challenging to separate task structure from perceptual variability. We introduce Grounded Scene-graph Reasoning (GSR), a structured reasoning paradigm that explicitly models world-state evolution as transitions over semantically grounded scene graphs. By reasoning step-wise over object states and spatial relations, rather than directly mapping perception to actions, GSR enables explicit reasoning about action preconditions, consequences, and goal satisfaction in a physically grounded space. To support learning such reasoning, we construct Manip-Cognition-1.6M, a large-scale dataset that jointly supervises world understanding, action planning, and goal interpretation. Extensive evaluations across RLBench, LIBERO, GSR-benchmark, and real-world robotic tasks show that GSR significantly improves zero-shot generalization and long-horizon task completion over prompting-based baselines. These results highlight explicit world-state representations as a key inductive bias for scalable embodied reasoning.

cs.RO↗

arXiv2025

FastTrack: a fast method to evaluate mass transport in solid leveraging universal machine learning interatomic potential

We introduce a rapid, accurate framework for computing atomic migration barriers in crystals by combining universal machine learning force fields (MLFFs) with 3D potential energy surface sampling and interpolation. Our method suppresses periodic self interactions via supercell expansion, builds a continuous PES from MLFF energies on a spatial grid, and extracts minimum energy pathways without predefined NEB images. Across twelve benchmark electrode and electrolyte materials including LiCoO2, LiFePO4, and LGPS our MLFF-derived barriers lie within tens of meV of DFT and experiment, while achieving ~10^2 x speedups over DFT-NEB. We benchmark GPTFF, CHGNet, and MACE, show that fine-tuning on PBE/PBE+U data further enhances accuracy, and provide an open-source package for high-throughput materials screening and interactive PES visualization.

cond-mat.mtrl-sci↗

arXiv2024

Developing Smart MAVs for Autonomous Inspection in GPS-denied Constructions

Smart Micro Aerial Vehicles (MAVs) have transformed infrastructure inspection by enabling efficient, high-resolution monitoring at various stages of construction, including hard-to-reach areas. Traditional manual operation of drones in GPS-denied environments, such as industrial facilities and infrastructure, is labour-intensive, tedious and prone to error. This study presents an innovative framework for smart MAV inspections in such complex and GPS-denied indoor environments. The framework features a hierarchical perception and planning system that identifies regions of interest and optimises task paths. It also presents an advanced MAV system with enhanced localisation and motion planning capabilities, integrated with Neural Reconstruction technology for comprehensive 3D reconstruction of building structures. The effectiveness of the framework was empirically validated in a 4,000 square meters indoor infrastructure facility with an interior length of 80 metres, a width of 50 metres and a height of 7 metres. The main structure consists of columns and walls. Experimental results show that our MAV system performs exceptionally well in autonomous inspection tasks, achieving a 100\% success rate in generating and executing scan paths. Extensive experiments validate the manoeuvrability of our developed MAV, achieving a 100\% success rate in motion planning with a tracking error of less than 0.1 metres. In addition, the enhanced reconstruction method using 3D Gaussian Splatting technology enables the generation of high-fidelity rendering models from the acquired data. Overall, our novel method represents a significant advancement in the use of robotics for infrastructure inspection.

cs.RO↗

arXiv2024

Accurate Cutting-point Estimation for Robotic Lychee Harvesting through Geometry-aware Learning

Accurately identifying lychee-picking points in unstructured orchard environments and obtaining their coordinate locations is critical to the success of lychee-picking robots. However, traditional two-dimensional (2D) image-based object detection methods often struggle due to the complex geometric structures of branches, leaves and fruits, leading to incorrect determination of lychee picking points. In this study, we propose a Fcaf3d-lychee network model specifically designed for the accurate localisation of lychee picking points. Point cloud data of lychee picking points in natural environments are acquired using Microsoft's Azure Kinect DK time-of-flight (TOF) camera through multi-view stitching. We augment the Fully Convolutional Anchor-Free 3D Object Detection (Fcaf3d) model with a squeeze-and-excitation(SE) module, which exploits human visual attention mechanisms for improved feature extraction of lychee picking points. The trained network model is evaluated on a test set of lychee-picking locations and achieves an impressive F1 score of 88.57%, significantly outperforming existing models. Subsequent three-dimensional (3D) position detection of picking points in real lychee orchard environments yields high accuracy, even under varying degrees of occlusion. Localisation errors of lychee picking points are within 1.5 cm in all directions, demonstrating the robustness and generality of the model.

cs.RO↗

arXiv2024

Exploring Accurate 3D Phenotyping in Greenhouse through Neural Radiance Fields

Accurate collection of plant phenotyping is critical to optimising sustainable farming practices in precision agriculture. Traditional phenotyping in controlled laboratory environments, while valuable, falls short in understanding plant growth under real-world conditions. Emerging sensor and digital technologies offer a promising approach for direct phenotyping of plants in farm environments. This study investigates a learning-based phenotyping method using the Neural Radiance Field to achieve accurate in-situ phenotyping of pepper plants in greenhouse environments. To quantitatively evaluate the performance of this method, traditional point cloud registration on 3D scanning data is implemented for comparison. Experimental result shows that NeRF(Neural Radiance Fields) achieves competitive accuracy compared to the 3D scanning methods. The mean distance error between the scanner-based method and the NeRF-based method is 0.865mm. This study shows that the learning-based NeRF method achieves similar accuracy to 3D scanning-based methods but with improved scalability and robustness.

cs.CV↗

arXiv2024

Pheno-Robot: An Auto-Digital Modelling System for In-Situ Phenotyping in the Field

Accurate reconstruction of plant models for phenotyping analysis is critical for optimising sustainable agricultural practices in precision agriculture. Traditional laboratory-based phenotyping, while valuable, falls short of understanding how plants grow under uncontrolled conditions. Robotic technologies offer a promising avenue for large-scale, direct phenotyping in real-world environments. This study explores the deployment of emerging robotics and digital technology in plant phenotyping to improve performance and efficiency. Three critical functional modules: environmental understanding, robotic motion planning, and in-situ phenotyping, are introduced to automate the entire process. Experimental results demonstrate the effectiveness of the system in agricultural environments. The pheno-robot system autonomously collects high-quality data by navigating around plants. In addition, the in-situ modelling model reconstructs high-quality plant models from the data collected by the robot. The developed robotic system shows high efficiency and robustness, demonstrating its potential to advance plant science in real-world agricultural environments.

cs.RO↗

arXiv2023

Developing Flying Explorer for Autonomous Digital Modelling in Wild Unknowns

This work presents an innovative solution for robotic odometry, path planning and exploration in wild unknown environments, focusing on digital modelling. The approach uses a minimum cost formulation with pseudo-randomly generated objectives, integrating multi-path planning and evaluation, with emphasis on full coverage of unknown maps based on feasible boundaries of interest. The evaluation carried out on a robotic platform with a lightweight 3D LiDAR sensor model, assesses the consistency and efficiency in exploring completely unknown subterranean-like areas. The algorithm allows for dynamic changes to the desired target and behaviour. At the same time, the paper details the design of AREX, highlighting its robust localisation, mapping and efficient exploration target selection capabilities, with a focus on continuity in exploration direction for increased efficiency and reduced odometry errors. The real-time, high-precision environmental perception module is identified as critical for accurate obstacle avoidance and exploration boundary identification.

cs.RO↗

arXiv2023

High-fidelity 3D Reconstruction of Plants using Neural Radiance Field

Accurate reconstruction of plant phenotypes plays a key role in optimising sustainable farming practices in the field of Precision Agriculture (PA). Currently, optical sensor-based approaches dominate the field, but the need for high-fidelity 3D reconstruction of crops and plants in unstructured agricultural environments remains challenging. Recently, a promising development has emerged in the form of Neural Radiance Field (NeRF), a novel method that utilises neural density fields. This technique has shown impressive performance in various novel vision synthesis tasks, but has remained relatively unexplored in the agricultural context. In our study, we focus on two fundamental tasks within plant phenotyping: (1) the synthesis of 2D novel-view images and (2) the 3D reconstruction of crop and plant models. We explore the world of neural radiance fields, in particular two SOTA methods: Instant-NGP, which excels in generating high-quality images with impressive training and inference speed, and Instant-NSR, which improves the reconstructed geometry by incorporating the Signed Distance Function (SDF) during training. In particular, we present a novel plant phenotype dataset comprising real plant images from production environments. This dataset is a first-of-its-kind initiative aimed at comprehensively exploring the advantages and limitations of NeRF in agricultural contexts. Our experimental results show that NeRF demonstrates commendable performance in the synthesis of novel-view images and is able to achieve reconstruction results that are competitive with Reality Capture, a leading commercial software for 3D Multi-View Stereo (MVS)-based reconstruction. However, our study also highlights certain drawbacks of NeRF, including relatively slow training speeds, performance limitations in cases of insufficient sampling, and challenges in obtaining geometry quality in complex setups.

cs.CV↗

arXiv2023

A Novel Perception and Semantic Mapping Method for Robot Autonomy in Orchards

Agricultural robots must navigate challenging dynamic and semi-structured environments. Recently, environmental modeling using LiDAR-based SLAM has shown promise in providing highly accurate geometry. However, how this chaotic environmental information can be used to achieve effective robot automation in the agricultural sector remains unexplored. In this study, we propose a novel semantic mapping and navigation framework for achieving robotic autonomy in orchards. It consists of two main components: a semantic processing module and a navigation module. First, we present a novel 3D detection network architecture, 3D-ODN, which can accurately process object instance information from point clouds. Second, we develop a framework to construct the visibility map by incorporating semantic information and terrain analysis. By combining these two critical components, our framework is evaluated in a number of key horticultural production scenarios, including a robotic system for in-situ phenotyping and daily monitoring, and a selective harvesting system in apple orchards. The experimental results show that our method can ensure high accuracy in understanding the environment and enable reliable robot autonomy in agricultural environments.

cs.RO↗

arXiv2022

Accurate Fruit Localisation for Robotic Harvesting using High Resolution LiDAR-Camera Fusion

Accurate depth-sensing plays a crucial role in securing a high success rate of robotic harvesting in natural orchard environments. Solid-state LiDAR (SSL), a recently introduced LiDAR technique, can perceive high-resolution geometric information of the scenes, which can be potential utilised to receive accurate depth information. Meanwhile, the fusion of the sensory information from LiDAR and camera can significantly enhance the sensing ability of the harvesting robots. This work introduces a LiDAR-camera fusion-based visual sensing and perception strategy to perform accurate fruit localisation for a harvesting robot in the apple orchards. Two SOTA extrinsic calibration methods, target-based and targetless-based, are applied and evaluated to obtain the accurate extrinsic matrix between the LiDAR and camera. With the extrinsic calibration, the point clouds and color images are fused to perform fruit localisation using a one-stage instance segmentation network. Experimental shows that LiDAR-camera achieves better quality on visual sensing in the natural environments. Meanwhile, introducing the LiDAR-camera fusion largely improves the accuracy and robustness of the fruit localisation. Specifically, the standard deviations of fruit localisation by using LiDAR-camera at 0.5 m, 1.2 m, and 1.8 m are 0.245, 0.227, and 0.275 cm respectively. These measurement error is only one one fifth of that from Realsense D455. Lastly, we have attached our visualised point cloud to demonstrate the highly accurate sensing method.

cs.RO↗

arXiv2021

Soft Robotic Finger with Variable Effective Length enabled by an Antagonistic Constraint Mechanism

Compared to traditional rigid robotics, soft robotics has attracted increasing attention due to its advantages as compliance, safety, and low cost. As an essential part of soft robotics, the soft robotic gripper also shows its superior while grasping the objects with irregular shapes. Recent research has been conducted to improve its grasping performance by adjusting the variable effective length (VEL). However, the VEL achieved by multi-chamber design or tunable stiffness shape memory material requires complex pneumatic circuit design or a time-consuming phase-changing process. This work proposes a fold-based soft robotic actuator made from 3D printed filament, NinjaFlex. It is experimentally tested and represented by the hyperelastic model. Mathematic and finite element modelling is conducted to study the bending behaviour of the proposed soft actuator. Besides, an antagonistic constraint mechanism is proposed to achieve the VEL, and the experiments demonstrate that better conformity is achieved. Finally, a two-mode gripper is designed and evaluated to demonstrate the advances of VEL on grasping performance.

cs.RO↗

arXiv2021

Geometry-Aware Fruit Grasping Estimation for Robotic Harvesting in Orchards

Field robotic harvesting is a promising technique in recent development of agricultural industry. It is vital for robots to recognise and localise fruits before the harvesting in natural orchards. However, the workspace of harvesting robots in orchards is complex: many fruits are occluded by branches and leaves. It is important to estimate a proper grasping pose for each fruit before performing the manipulation. In this study, a geometry-aware network, A3N, is proposed to perform end-to-end instance segmentation and grasping estimation using both color and geometry sensory data from a RGB-D camera. Besides, workspace geometry modelling is applied to assist the robotic manipulation. Moreover, we implement a global-to-local scanning strategy, which enables robots to accurately recognise and retrieve fruits in field environments with two consumer-level RGB-D cameras. We also evaluate the accuracy and robustness of proposed network comprehensively in experiments. The experimental results show that A3N achieves 0.873 on instance segmentation accuracy, with an average computation time of 35 ms. The average accuracy of grasping estimation is 0.61 cm and 4.8$^{\circ}$ in centre and orientation, respectively. Overall, the robotic system that utilizes the global-to-local scanning and A3N, achieves success rate of harvesting ranging from 70\% - 85\% in field harvesting experiments.

cs.RO↗

arXiv2021

A Tactile-enabled Grasping Method for Robotic Fruit Harvesting

In the robotic crop harvesting environment, foreign objects intrusion in the gripper workspace is frequently occurring and unignorable, however, rarely addressed. This paper presents a novel intelligent robotic grasping method capable of handling obstacle interference, which is the first of its kind in the literature. The proposed method combines deep learning algorithms with low-cost tactile sensing hardware on a multi-DoF soft robotic gripper. Through experimental validations, the proposed method demonstrated promising performance in distinguishing various grasping scenarios. The 4-finger independently controlled gripper presented outstanding adaptability to handle various picking scenarios. The overall performance of this work indicated great potential for solving the robotic fruit harvesting challenges.

cs.RO↗

arXiv2020

Real-Time Fruit Recognition and Grasping Estimation for Autonomous Apple Harvesting

In this research, a fully neural network based visual perception framework for autonomous apple harvesting is proposed. The proposed framework includes a multi-function neural network for fruit recognition and a Pointnet grasp estimation to determine the proper grasp pose to guide the robotic execution. Fruit recognition takes raw input of RGB images from the RGB-D camera to perform fruit detection and instance segmentation, and Pointnet grasp estimation take point cloud of each fruit as input and output the prediction of grasp pose for each of fruits. The proposed framework is validated by using RGB-D images collected from laboratory and orchard environments, a robotic grasping test in a controlled environment is also included in the experiments. Experimental shows that the proposed framework can accurately localise and estimate the grasp pose for robotic grasping.

cs.CV↗