arXivSub Start free trial

IROS 2025 Papers with AI Summaries

IEEE/RSJ International Conference on Intelligent Robots and Systems · 1984 papers

16 Ways to Gallop: Energetics and Body Dynamics of High-Speed Quadrupedal Gaits

Yasser G. Alqaham, Zhenyu Gan

OptimizationRobotic Intelligence

🎯 What it does: Systematically analyze and classify different flight phases of the horse-shoe gait, establish a hybrid dynamic model using the A1 robot, apply trajectory optimization to minimize energy consumption, and verify it using a QP controller in Gazebo

2.5D Object Mapping using Gaussian Processes for Robot Navigation

Erdem Toraman, Emre Özkan

Robotic IntelligencePoint Cloud

🎯 What it does: Propose a probabilistic framework for 2.5D object modeling and information-driven motion planning based on Gaussian processes, implemented using LiDAR measurements.

3-D Multiple Sound Source Localization Based on a Five-Element Microphone Array

Yizhen Qiu, Zhiyao Huang

Audio

🎯 What it does: A 3-D multi-source localization method based on a five-microphone array is proposed. It first uses independent vector analysis (IVA) to separate source signals, then achieves precise localization of each source through four time difference of arrival (TDOA) values.

360Recon: An Accurate Reconstruction Method based on Depth Fusion from 360 Images

Zhongmiao Yan, Ling Pei

GenerationDepth EstimationConvolutional Neural NetworkImage

🎯 What it does: Proposed a multi-view stereo reconstruction algorithm called 360Recon for 360° panoramic images

3D Gaussian Splatting for Fine-Detailed Surface Reconstruction in Large-Scale Scene

Shihan Chen, Ran Duan

Neural Radiance FieldGaussian SplattingImage

🎯 What it does: Utilizing the 3D Gaussian Splatting method, combined with a coarse-to-fine reconstruction strategy, scene-adaptive partitioning and sub-scene refinement, decoupled appearance model, transient mask model, extended multi-view constraints, and single-view texture smoothing regularization, achieving high-detail surface reconstruction for large-scale scenes under full-size image supervision.

3D Hierarchical Panoptic Segmentation in Real Orchard Environments Across Different Sensors

Matteo Sodano (University of Bonn), C. Stachniss (University of Bonn)

SegmentationPoint CloudAgriculture Related

🎯 What it does: Proposed a hierarchical panoramic segmentation method for 3D data in apple orchards, which can simultaneously perform semantic segmentation, stem and fruit instance segmentation, as well as tree (stem + fruit) instance segmentation; the method can identify individual plants, fruits, and stems, and calculate the number of fruits per tree.

3D Plant Root Skeleton Detection and Extraction

Jiakai Lin, Guoyu Lu

Object DetectionImageAgriculture Related

🎯 What it does: Propose an efficient method for extracting the 3D skeleton of plant roots from a small number of images, including lateral root detection and matching, triangulation, and main-lateral root fusion.

3D Vision-tactile Reconstruction from Infrared and Visible Images for Robotic Fine-grained Tactile Perception

Yuankai Lin, Hua Yang

Depth EstimationRobotic IntelligenceImage

🎯 What it does: Developed GelSplitter3D, which extends the imaging channel through a prism and near-infrared camera, proposed a photometric stereo neural network based on CAD-generated normal ground truth, and designed a normal integration method utilizing depth prior boundary constraints.

3D-AMTA: Occlusion-Aware Real-Time 3D Hand Pose Estimation with Auto Mask and Token-Specific Attention

Dongfang Zhao, Donghoon Kim

Pose EstimationTransformerImage

🎯 What it does: Proposes the 3D-AMTA framework based on Transformer, combining Auto Mask and Token-specific Attention to achieve occlusion-aware 3D hand pose estimation from a single RGB image.

3D-MoRe: Unified Modal-Contextual Reasoning for Embodied Question Answering

Rongtao Xu, Shibiao Xu

GenerationData SynthesisTransformerLarge Language ModelVision Language ModelVision-Language-Action ModelTextMultimodalityPoint Cloud

🎯 What it does: Generate a large-scale 3D-language dataset for question answering and object description generation

3D-UnOutDet: A Fast and Efficient Unsupervised Snow Removal Algorithm for 3D LiDAR Point Clouds

Abu Mohammed Raisuddin, E. Aksoy

RestorationAutonomous DrivingPoint Cloud

🎯 What it does: Proposed a fast and memory-efficient unsupervised statistical method combined with an unsupervised deep learning model for snow removal in 3D LiDAR point clouds.

3DS-SLAM: A 3D Object Detection based Semantic SLAM towards Dynamic Indoor Environments

G. S. Krishna, S. Baidya

Object DetectionRobotic IntelligenceTransformerSimultaneous Localization and MappingPoint Cloud

🎯 What it does: Proposes a 3D semantic SLAM system called 3DS-SLAM for dynamic indoor environments

3DWSNet: A Novel 3D Wavelet Spiking Neural Network for Event-based Action Recognition

Junkang Fang, Jianqin Yin

RecognitionSpiking Neural Network

🎯 What it does: Proposed a 3D Waveform Synapse Neural Network (3DWSNet) for action recognition using event cameras.

4D-ROLLS: 4D Radar Occupancy Learning via LiDAR Supervision

Ruihan Liu, Yunjiang Lou

Autonomous DrivingPoint Cloud

🎯 What it does: Designed and implemented a 4D radar occupancy estimation method based on LiDAR point cloud supervision, proposing pseudo LiDAR label generation and multi-stage supervision, leveraging LiDAR occupancy maps for model alignment and fine-tuning.

6-DoF Object Tracking with Event-based Optical Flow and Frames

Zhichao Li, Lorenzo Natale

Object TrackingPose EstimationOptical FlowImage

🎯 What it does: Propose a 6-DoF pose tracking method that combines event camera optical flow with RGB global pose estimator, utilizing event optical flow to track object velocity and fusing velocity with low-frequency pose estimation to achieve high-speed motion tracking.

A 4D Radar Camera Extrinsic Calibration Tool Based on 3D Uncertainty Perspective N Points

Chuan Cao, Jingchuan Wang

Pose EstimationAutonomous DrivingImagePoint Cloud

🎯 What it does: Propose a 4D radar camera external calibration tool based on a 3D uncertainty perspective, which explicitly models radar spherical coordinate noise propagation and compensates for non-zero error expectations in coordinate transformations using the 3DUPnP algorithm

A Bio-inspired Robotic Electric Ray Design of Multimodal Locomotion with Grasping Function

Yuyang Mo, Yong Zhong

Robotic Intelligence

🎯 What it does: Designed a bio-inspired electric ray robot that integrates BCF-mode tail fin and MPF-mode pectoral fin to achieve multimodal locomotion and grasping functions.

A Bio-inspired Spherical Soft Magnetic Millirobot for Gastrointestinal Applications

Yulin Li, Liu Yang

Robotic IntelligenceDrug DiscoveryBiomedical Data

🎯 What it does: Proposed a spherical soft magnetic microrobot (S2M2), verified in ex vivo pig stomach experiments for its movement, adhesion capability, and pressure-based drug delivery function on gastric mucosal surfaces.

A Bio-inspired Stiffness-programmable Robotic Flexible Joint Based on Electro-adhesive Clutches

Yongxian Ma, Xiaozheng Li

Robotic Intelligence

🎯 What it does: Proposed and implemented a bio-inspired variable stiffness flexible joint (SPRFJ) based on electro-adhesive clutches, and verified its performance through experiments.

A Bioinspired Framework for Person Detection and Tracking using Events and Frames on Flapping-Wing Aerial Robots

R. Tapia, Aníbal Ollero

Object DetectionObject TrackingRobotic IntelligenceConvolutional Neural NetworkImage

🎯 What it does: Proposes a bio-inspired framework for human detection and tracking in flapping-wing UAVs, combining event-level blob tracker with frame-based deep neural network detector;

A Bionic Robotic Hand Designed with Multiple Grasping Modes and Magnetic-tactile Perception

Shixian Wang, Bin Fang

Robotic Intelligence

🎯 What it does: Designed and implemented a multi-modal robotic hand integrating magnetic silicone skin, rigid skeleton, and suction cups, supporting three grasping modes: encapsulation, parallel, and suction grasping.

A Coarse-to-Fine Approach to Multi-Modality 3D Occupancy Grounding

Zhan Shi, Jianke Zhu

Depth EstimationAutonomous DrivingVision Language ModelMultimodalityBenchmark

🎯 What it does: Proposes a 3D occupancy ground benchmark and an end-to-end multi-modal model called GroundingOcc for precisely locating and predicting occupied voxels in outdoor scenes based on natural language descriptions.

A Compact Dual-Mode Twisting Retraction Device for Endoscopic Submucosal Dissection

Haley Mayer, Eric D. Diller

🎯 What it does: Designed and manufactured a compact dual-mode wireless traction device for endoscopic submucosal dissection (ESD), capable of achieving high tensile force and adjustable traction through magnetic drive;

A Compact Robotic Wrist with Embedded Torque Sensing for Peg-in-Hole Tasks

Yi-Shian Tsai, Chao-Chieh Lan

Robotic Intelligence

🎯 What it does: Designed and experimentally verified a compact robotic wrist with embedded torque sensing for peg-in-hole tasks.

A Compliant Tube for Series Elastic Actuators to Generate High Output Torque

Chun-Hung Huang, Chao-Chieh Lan

Robotic Intelligence

🎯 What it does: This paper proposes a novel spiral flexible tube used as a spring in series elastic actuators (SEA), which can significantly increase output torque while maintaining a compact size, and demonstrates zero-torque and high-torque motion through a prototype.

A Comprehensive Evaluation of LiDAR Odometry Techniques

Easton R. Potokar, Michael Kaess

Pose EstimationAutonomous DrivingSimultaneous Localization and MappingPoint CloudBenchmark

🎯 What it does: Summarized and experimentally evaluated technical components in the LiDAR localization (LO) pipeline, covering multiple environments, LiDAR types, and vehicle motions.

A Convex Formulation of Material Points and Rigid Bodies with GPU-Accelerated Async-Coupling for Interactive Simulation

Chang Yu, Xuchen Han

Optimization

🎯 What it does: Proposed a new convex optimization formulation that weakly couples MPM with rigid body dynamics through frictional contact, and implemented GPU parallel optimization to achieve interactive simulation.

A Cooperative Contactless Object Transport with Acoustic Robots

Narsimlu Kemsaram, S. Subramanian

Robotic IntelligenceUltrasoundAudio

🎯 What it does: Proposed an acoustic-based robotic system that achieves collaborative and independent control for contactless object transportation in air.

A Crab-Inspired Soft Gripper with Single-Finger Dexterous Grasping Capabilities

Yunce Zhang, Rui Song

Robotic Intelligence

🎯 What it does: This paper proposes and implements a single-finger flexible gripper inspired by the structure of crab claws, and verifies its bending model and grasping performance through experiments.

A Da Vinci Open Spina Bifida Suturing Simulator with Continuum Tools for Surgeon Skills Training

N. Nimal, T. Looi

Robotic IntelligenceBiomedical Data

🎯 What it does: Developed a Da Vinci open spina bifida repair simulator with continuous tools for surgeon training.

A Deep Learning-Driven Autonomous System for Retinal Vein Cannulation: Validation Using a Chicken Embryo Model

Yi Wang, I. Iordachita

Depth EstimationRobotic IntelligenceBiomedical Data

🎯 What it does: Developed an automated robotic system integrating top-view microscopy and B-scan optical coherence tomography (OCT) to achieve precise depth perception and needle tip navigation for retinal vein cannulation.

A Deep Reinforcement Learning based End-to-End Control Framework for Lower Limb Exoskeletons with Smooth Movement Transitions

Minsu Kim, Jaeheung Park

Robotic IntelligenceReinforcement Learning

🎯 What it does: Proposed an end-to-end control framework based on deep reinforcement learning for lower-limb exoskeletons to achieve smooth transitions between different motion modes

A Dual Tiltrotor UAV with Foldable Wings for Passive Perching and Belly/Back Takeoff

Luyao Wang, Xiang He

🎯 What it does: Propose a dual-tilt-rotor UAV equipped with foldable wings and a precisely positioned center of gravity to achieve passive hovering and multimodal flight.

A Dual-Arm Shared Control Framework Integrating Sub-goals and Predicted Trajectories for Asymmetric Tasks

Zhixiong Wang, Fang Gao

Robotic Intelligence

🎯 What it does: Proposed a dual-arm shared control framework integrating a subgoal generation module and a trajectory prediction module to enhance fine-grained collaboration and motion smoothness in asymmetric tasks.

A Fast-moving Underwater Wall-climbing Robotic Fish Inspired by Rock-climbing Fish

Hengshen Qin, Lianqing Liu

Robotic Intelligence

🎯 What it does: Based on the stretchable friction characteristics of rockfish, an adhesion module, pectoral and abdominal fin oscillation modules, and a novel high-frequency line-driven tail fin were designed, constructing a bio-inspired robotic fish capable of adhesion-sliding movement on underwater walls.

A Flexible Bending Sensor Based on C-Shaped FBG Array for Curvature and Gesture Recognition

Baijin Mao, Juntian Qu

Recognition

🎯 What it does: Proposed and implemented a C-shaped flexible bending sensor with fiber Bragg grating (FBG) arrays embedded in a 2 mm silicone layer for bending and gesture recognition.

A Gait Phase Detection and Gait Spatio-temporal Features Extraction Method Based on the Inertial Measurement Unit*

Shuai Fan, Peng Chen

Time SeriesBiomedical Data

🎯 What it does: Propose a home-based gait feature extraction method using inertial measurement units (IMUs), calculating walking distance and speed via double integration, determining step count through local maximum peak detection, and calculating stance and swing phases using local minimum valley detection.

A Human Reasons-based Supervision Framework for Ethical Decision-Making in Automated Vehicles

Lucas Elbert Suryana, B. Arem

Autonomous Driving

🎯 What it does: Developed a supervised framework based on human reasoning that can detect inconsistencies between autonomous vehicle behaviors and expected human rationale, and trigger trajectory replanning when necessary to enhance safety, efficiency, and compliance.

A Hybrid Mapping Method: Balancing Efficiency and Intuitiveness in Lateral Teleoperation

Yuwei Xie, Gaofeng Li

Computational Efficiency

🎯 What it does: Proposed a hybrid mapping method to achieve fast and intuitive lateral teleoperation.

A Kinematics Constrained Convex Optimal Trajectory Generation Method for Robotic-assisted Flexible Needle

Fan Ren, Xiangyu Wang

OptimizationRobotic Intelligence

🎯 What it does: Proposes a kinematic constraint convex optimal trajectory generation method for flexible needles with higher maximum curvature to avoid collisions with vital organs during surgery.

A Learning Quasi-stiffness Control Framework of a Powered Transfemoral Prosthesis for Adaptive Speed and Incline Walking

Teng Ma, Chenglong Fu

Robotic Intelligence

🎯 What it does: Proposed a task-adaptive quasi-stiffness control framework for powered transfemoral prosthetic legs, designed for walking at varying speeds and slopes.

A light-controlled micromixer using optoelectronic tweezers

Feng Yan, Na Liu

Physics Related

🎯 What it does: Developed a flexible and effective micro stirrer based on opto-electro-thermal (OET) plating wire, utilizing asymmetric induced charge electro-osmosis (ICEO) and dielectrophoresis (DEP) phenomena on NdFeB micro-scale anisotropic particles.

A Lightweight 3-axis Permanent Magnetic Sponge-based Self-Adapting Tactile Sensor

Yushi Wang, Mitsuhiro Kamezaki

🎯 What it does: Designed, fabricated, and characterized a lightweight triaxial permanent magnetic sponge adaptive tactile sensor.

A Mole-inspired Incisor-Burrowing Robotic Platform for Planetary Exploration

Ran Xu, Xilun Ding

Robotic Intelligence

🎯 What it does: This paper designs and implements a robot platform named MIRP that mimics the incisor digging behavior of naked mole-rats, combining quadrupedal locomotion with digging mechanisms, and verifies its mobility and digging performance in a simulated lunar soil environment.

A Monocular Vision-based Robotic Arm Teleoperation Method for Human Arm Configuration Imitation

Jindong Xiang, Mengtang Li

Pose EstimationRobotic IntelligenceImage

🎯 What it does: Propose a remote control system based on a monocular RGB camera, which uses computer vision to capture human movements and maps human joint configurations to various robotic arm structures through kinematic mapping, achieving end-effector tracking and imitation of human joint configurations.

A Multi-Modal Benchmark for Long-Range Depth Evaluation in Adverse Weather Conditions

Stefanie Walz, Felix Heide

Depth EstimationMultimodalityPoint CloudBenchmark

🎯 What it does: Proposes a multi-modal long-range depth evaluation benchmark covering clear, foggy, and rainy scenes up to 200 meters, and obtains dense ground truth through multi-modal sensors and millimeter-level digital twins.

A Multi-modal Hand Imitation Dataset for Dexterous Hand

Shaochen Wang, Beihao Xia

Pose EstimationRobotic IntelligenceNeural Radiance FieldImageMultimodalityPoint Cloud

🎯 What it does: Built and released a multimodal Dex Hand Dataset integrating visual, point cloud, and kinematic data, aligning human and robotic hand poses in a shared canonical space using neural rendering and kinematic optimization to achieve fine-grained hand imitation learning.

A Multi-Sensor Fusion Approach for Rapid Orthoimage Generation in Large-Scale UAV Mapping

Jialei He, Jie Yuan

GenerationImageMultimodalityPoint Cloud

🎯 What it does: Utilizes a multi-sensor drone system (GPS, IMU, 4D millimeter-wave radar, and camera) to rapidly generate large-scale aerial orthophotos through multi-sensor data.

A Multi-Task Learning System for Composites Defect Segmentation and Classification with TacRoller

Xiaolong Li, Chenguang Yang

ClassificationSegmentationConvolutional Neural NetworkImage

🎯 What it does: Designed the TacRoller tactile sensing roller for automatically capturing tactile images of composite material surface defects, and collected 2,744 samples containing ripple, foreign objects, broken fibers, voids, and healthy textures; subsequently proposed the MOGA-U-Net model to complete defect segmentation and multi-class classification tasks.

A Multimodal Neural Network for Recognizing Subjective Self-Disclosure Towards Social Robots

Henry Powell, Emily S. Cross

ClassificationRecognitionRobotic IntelligenceVideo

🎯 What it does: Developed a multimodal attention network based on an emotion recognition model for identifying human subjective self-disclosure toward social robots.

A Natural Human-Robot Interaction System for Teleoperation Based on Noncontact Haptic Feedback

Letian Wei, Mengchu Zhou

Robotic IntelligenceUltrasound

🎯 What it does: Developed a natural human-computer interaction system based on ultrasound haptic feedback, achieving low-latency, high-fidelity teleoperation

A New Concept for Reconstruction of Volumetric Muscle Loss Injuries Using Spatial Robotic Embedded Bioprinting: A Feasibility Study

Omid Rezayof, F. Alambeigi

Robotic IntelligenceBiomedical Data

🎯 What it does: Proposed and implemented a novel concept for reconstructing volumetric muscle loss (VML) trauma—spatial robotic embedded bioprinting—and achieved three-dimensional printing of muscle fiber structures through a framework combining a robotic arm with a custom syringe.

A New Unsupervised Infrared and Visible Image Fusion Method Based on Salient Object Segmentation under Poor Illumination

Zheng-Feng Wang, Zhiyao Huang

SegmentationImage

🎯 What it does: Proposed an unsupervised infrared and visible light image fusion method based on salient object segmentation, which includes improved superpixel segmentation, density peaks-based clustering, improved GrabCut strategy, and partition fusion strategy.

A No-code Approach for Intuitive Robot Programming for Process-Aligned Surface Processing

Jayanto Halim, Steffen Ihlenfeldt

Robotic IntelligenceVision-Language-Action Model

🎯 What it does: Proposed a process-aligned robot path generation method for surface machining

A Noise-Robust Turn-Taking System for Real-World Dialogue Robots: A Field Experiment

K. Inoue, Tatsuya Kawahara

RecognitionRobotic IntelligenceTransformerAudio

🎯 What it does: Proposed and evaluated a Transformer-based noise-robust speech activity projection (VAP) model to enhance the real-time interaction smoothness of chatbots

A Novel Aerial-Aquatic Locomotion Robot with Variable Stiffness Propulsion Module

Junzhe Hu, Janet Dong

Robotic Intelligence

🎯 What it does: Designed and manufactured a novel aerial-aquatic amphibious motion robot (AALR), integrating a variable stiffness propulsion module (VSPM) on the quadrotor arm to achieve continuous motion in both air and underwater environments.

A Novel Effective Loop Gait and Stabilizing Morphology Parameterization in Snake Robots

Chaoquan Tang, Peng Li

OptimizationRobotic Intelligence

🎯 What it does: Propose the Loop gait to enhance the locomotion speed and energy efficiency of serpentine robots, and develop a static stability model for parameter optimization

A novel event-based structured light system for high-precision and high-speed depth sensing

Gong Su, Wei Xi

Depth EstimationComputational Efficiency

🎯 What it does: Proposed an event-based depth perception system combining line laser scanning, polygon mirror scanners, and an efficient plane-ray intersection reconstruction pipeline.

A Novel Gripper with Semi-Peaucellier Linkage and Idle-Stroke Mechanism for Linear Pinching and Self-Adaptive Grasping *

Haokai Ding, Wenzeng Zhang

Robotic Intelligence

🎯 What it does: Designed and implemented an SPD gripper with linear parallel clamping and adaptive grasping capabilities, and verified its performance through experiments.

A Novel Large-Scale Collaborative Mapping Framework with Heterogeneous Point Clouds for Aerial-Ground Robots

Shuang Luan, Zhuang Yan

Autonomous DrivingOptimizationConvolutional Neural NetworkSimultaneous Localization and MappingPoint Cloud

🎯 What it does: Proposed a LiDAR-based aerial-ground robot collaborative mapping framework, including cross-source pose recognition, overlapping perception registration, and global graph optimization;

A Novel LiDAR Odometry Based on Surface Distributed Point Feature with Dual Feature Fusion

Jianjie Li, Zhiqiang Cao

Pose EstimationAutonomous DrivingSimultaneous Localization and MappingPoint Cloud

🎯 What it does: Propose a LiDAR odometry method based on Surface Distribution Points (SDP) features and dual feature fusion, applied in scan-to-map matching and fixed-lag smoothing

A Novel Robot Hand with Hoeckens Linkages and Soft Phalanges for Scooping and Self-Adaptive Grasping in Environmental Constraints

Wentao Guo, Wenzeng Zhang

Robotic Intelligence

🎯 What it does: Designed and verified a novel adaptive robotic hand, Hockens-A Hand, achieving three grasping modes: parallel pinch, asymmetric scooping, and enveloping grasp.

A Novel Strategy for Connectivity Maintenance and Recovery in Heterogeneous Multi-Robot Systems

Enrico Fiasché, Philippe Martinet

Robotic Intelligence

🎯 What it does: Developed a strategy that combines model predictive control (MPC) with general connectivity maintenance (GCM) to support real-time path adaptation while maintaining connectivity in multi-robot systems, and introduced a decentralized decision-making framework to enable dynamic recovery of disconnected members; when reconnection is impossible, the system adaptively adjusts tasks to continue execution.

A Novel Terrain Classification System with Planar ECT Sensor

Wenju Yang, Di Guo

ClassificationRobotic IntelligenceMultimodality

🎯 What it does: A novel terrain classification system was developed, utilizing a planar electrical capacitance tomography (ECT) sensor integrated into the bottom of a mobile robot to collect capacitance and IMU measurement data, which is then classified using a multi-layer perceptron network.

A Parallel Fuzzy Nonlinear ADRC Framework for Robotic Machining with Vibrations

X. Hao, Yaonan Wang

Robotic Intelligence

🎯 What it does: A parallel fuzzy nonlinear active disturbance rejection control (FNLADRC) strategy is proposed to improve the accuracy and robustness of robots during the machining of large complex parts.

A Partition-Learning-Selection-Augmentation (PLSA) Framework to Solve Forward Kinematics of Parallel Robots

Ruiqi Xiang, Mengtang Li

OptimizationRobotic IntelligenceConvolutional Neural NetworkTransformerPhysics Related

🎯 What it does: Propose a Partition-Learning-Selection-Augmentation (PLSA) framework to systematically address the multi-solution problem of forward kinematics in parallel robots, leveraging data partitioning, deep neural networks for parallel prediction of all feasible solutions, combined with a selection mechanism to pick the optimal solution and further refine it via the Newton-Raphson method;

A Patch-Based Transformer Method for Electrical Capacitance Tomography Image Reconstruction

Yuliang Wang, Di Guo

RestorationTransformerImage

🎯 What it does: Proposed a patch-based Transformer method for reconstructing capacitance imaging images.

A Physics-Informed Neural Network for the Calibration of Electromagnetic Navigation Systems

Pascal Ernst, Q. Boehler

Physics Related

🎯 What it does: Research and propose a physics-informed neural network for calibrating electromagnetic navigation systems, aiming to improve the accuracy of magnetic field and gradient predictions, and to determine the corresponding electromagnetic currents based on the desired field and gradient.

A Planning Framework for Stable Robust Multi-Contact Manipulation

Lin Yang, Domenico Campolo

OptimizationRobotic Intelligence

🎯 What it does: Propose a planning-and-optimization-based multi-contact stable robust manipulation framework, evaluate contact stability and robustness to sensor noise, and validate it on dual-arm planar insertion tasks and multi-manipulator multi-hole insertion tasks.

A Probabilistic Programming Approach to Intention Estimation in Human-Robot Teleoperated Assembly Tasks

Aolin Xu, Behzad Dariush

Robotic IntelligenceTransformer

🎯 What it does: Propose a probabilistic programming-based model for intent estimation in remote human-robot collaborative assembly tasks, including task estimation and action prediction.

A Real-Time Collision-Avoidance Motion Planner for Robot Soccer

A. Deogan, R. V. D. Molengraft

Robotic Intelligence

🎯 What it does: Implemented and evaluated a real-time collision avoidance motion planning algorithm for high dynamic environments

A Reconfigurable Manipulator with Schönflies and RCM Motions

Tianye Xu, Xilun Ding

Robotic Intelligence

🎯 What it does: Designed a reconfigurable robotic arm capable of performing Schönflies and RCM operations.

A Recursive Total Least Squares Solution for Bearing-Only Target Motion Analysis and Circumnavigation

Lin Li, Qingrui Zhang

Autonomous DrivingOptimizationComputational EfficiencySequential

🎯 What it does: Proposed a recursive total least squares (RTLS) method for online target localization and tracking using a mobile observer, and designed a surrounding controller to enhance system observability and estimation convergence;

A Rehabilitation Robot System to Enhance Proprioception with Physical and Virtual Simulation of Multi-terrain Scenarios

Liziyi Hao, Ningbo Yu

Robotic IntelligenceBiomedical Data

🎯 What it does: Developed a neural rehabilitation robot platform capable of simulating various terrains (such as uneven mountain paths, sand, and bumpy buses) in both physical and virtual environments to enhance patients' proprioception and assess their stability and recovery ability.

A Ribbed Hybrid Rigid-Flexible Tail with Graded Stiffness and Anisotropic Friction for Enhanced Robot Locomotion and Fall Damage Prevention

Pongsiri Borijindakul, P. Manoonpong

Robotic Intelligence

🎯 What it does: Proposed a hybrid rigid-flexible robot tail (HIFLEX) with gradient stiffness and anisotropic friction, and experimentally validated its crawling performance on rough and deformable slopes as well as its ability to protect against fall damage.

A Rigid-flexible Coupled Bionic Robotic Finger with Perception Decoupling and Slip Detection Capabilities

Yuyaocen Xiang, Juntian Qu

Robotic Intelligence

🎯 What it does: Developed a stiffness-flexibility coupling bio-inspired robotic finger, achieving perception decoupling and sliding detection functions.

A Robust Distributed Odometry for Mobile Robots with Steerable Wheels

Wang Xi, Jianping He

Robotic Intelligence

🎯 What it does: Proposes a distributed odometry framework named ICF-DO for mobile robots with steerable wheels, applicable to SWMR and multi-wheel collaborative robot systems.

A robust MLTD3 path planning algorithm in unknown environments

Tianqing Wen, Zhendong Sun

Robotic IntelligenceRecurrent Neural NetworkReinforcement Learning

🎯 What it does: Proposed a Multi-Layer Long Short-Term Memory Double Delayed Deep Deterministic Policy Gradient (MLTD3) algorithm for path planning of mobile robots in unknown environments.

A Robust Stereo Splatting SLAM System with Inertial-Legged Fusion

Zuowei Chen, Qing Shi

Robotic IntelligenceGaussian SplattingSimultaneous Localization and MappingImageBenchmark

🎯 What it does: Developed a robust stereo 3D Gaussian Splatting (3DGS) SLAM system for small quadruped robots (SaQu-SLAM), incorporating a lightweight network for dense initialization of 3D Gaussians, Kalman filter-based fusion of inertial and leg encoder data, and a mode switching mechanism to handle visual intermittency failures.

A Rubber-sheet Transformation Model for Personalized Human-Robot Proxemics

Fanta Camara, M. Obaid

Robotic IntelligenceTabular

🎯 What it does: Proposed a personalized method for representing human-machine proximity using a rubber band deformation model

A Safe and Convenient Feeding Assistive Robotic Based on Multi-modal Interaction Method

Jiahui Ding, Toshio Fukuda

Pose EstimationRobotic IntelligenceLarge Language ModelVision Language ModelMultimodality

🎯 What it does: Established a multimodal interactive feeding assistance robot system and proposed a safe and comfortable feeding method that combines visual recognition with LLM-based voice interaction.

A Safety-Adjusted Policy Optimization Algorithm and Application for Obstacle Avoidance in the Quadcopter

Gang Xia, Xiwang Dong

OptimizationRobotic IntelligenceReinforcement Learning

🎯 What it does: Proposed a dual-policy framework safety regulation strategy optimization algorithm (SAPO) to address safety constraint violations in reinforcement learning, and applied it to quadrotor obstacle avoidance navigation.

A Safety-Enhanced Autonomous Resection Method for Precision Laparoscopic Surgery amid Tissue Deformation

Yudong Shi, Xiaojian Li

Safty and PrivacyRobotic Intelligence

🎯 What it does: A safe enhanced autonomous resection method for laparoscopic surgery is proposed, including pre-planned cutting paths, navigation switching strategies based on a preview tracking mechanism, deep fusion navigation controller, safe withdrawal motion controller, inertial tracking mechanism, and a confidence generator that fuses two controllers. Its effectiveness is validated through simulation and physical experiments.

A Self-Sensing Phase-Change Buoyancy System for Miniaturized Deep-sea Robotics

Zonghao Zuo, Li Wen

Robotic Intelligence

🎯 What it does: A self-sensing phase-change buoyancy regulation system was developed for small deep-sea robots, capable of operating in extreme deep-sea environments.

A Sim-to-Real Transfer Framework for Enhancing Marine Vehicle Performance in Ocean Environments

Ze Zheng, Wenbo Xie

Domain AdaptationAutonomous DrivingReinforcement Learning

🎯 What it does: Built a simulation-to-real transfer framework for marine vehicles, combining high-fidelity data-driven dynamics models with multi-factor domain randomization, and evaluated path tracking tasks through a simulation-simulation platform.

A Simple Algebraic Solution for Estimating the Pose of a Camera from Planar Point Features

Tarek Bouazza, Claude Samson

Pose EstimationPoint Cloud

🎯 What it does: Propose a simple algebraic method to estimate the camera pose from at least four reference points with known coordinates on a planar target and line-of-sight measurements in the camera coordinate system.

A Simple Approach to Constraint-Aware Imitation Learning with Application to Autonomous Racing

Shengfan Cao, Francesco Borrelli

Autonomous DrivingReinforcement LearningImage

🎯 What it does: Propose a simple method to integrate safety constraints into the imitation learning objective

A Simple yet Effective Test-Time Adaptation for Zero-Shot Monocular Metric Depth Estimation

Rémi Marsal, David Filliat

Depth EstimationDomain AdaptationSimultaneous Localization and MappingPoint Cloud

🎯 What it does: Propose a method to recalibrate the predictions of Depth Anything to obtain metric depth by utilizing 3D points (low-resolution LiDAR or SfM+IMU) without fine-tuning.

A Situation-Aware Autonomous Camera Alignment for Enhanced Suturing in Robot-Assisted Minimally Invasive Surgery

Elisa Iovene, I. Nisky

Robotic IntelligenceBiomedical Data

🎯 What it does: Developed a context-aware autonomous camera alignment system based on real-time kinematic data to predict user intent, dynamically adjusting the camera position during simulated suturing tasks in robotic-assisted surgery.

A Skill-Based Hierarchical Framework with Dangerous Action Masking for Autonomous Navigation of Jumping Robots

Gangyang Li, Qing Shi

Robotic IntelligenceReinforcement Learning

🎯 What it does: Propose a skill-based hierarchical framework combined with danger action shielding for autonomous navigation in jumping robots

A Smooth Analytical Formulation of Collision Detection and Rigid Body Dynamics With Contact

O. Beker, Sylvain Calinon

Physics Related

🎯 What it does: Proposes an analytical method that jointly integrates collision detection and contact modeling, generating fully analytical and smooth (twice-differentiable) forward and inverse dynamics;

A Soft Active Surface Gripper for Safe In Hand Manipulation of Fragile Objects

Shen Xiang, Yang Yang

Robotic Intelligence

🎯 What it does: Designed and fabricated a soft active surface gripper for safely manipulating fragile objects within the hand.

A Soft Robot Attachment with Variable Stiffness Effector for Advanced Endoscopic Surgical Tasks

Zhangxi Zhou, George P. Mylonas

Robotic IntelligenceBiomedical Data

🎯 What it does: Designed and verified a soft cable-driven parallel robot with a variable stiffness end-effector for gastrointestinal endoscopic surgery, capable of navigating through a colon model in a soft state and increasing stiffness at lesion sites to collaborate with the endoscope for surgical procedures.

A Spatial Position-based Visual Servoing Obstacle-avoidable Shape Control Framework for An 11-DOF Hybrid Continuum Robot

P. Zhu, Kwok Wai Samuel Au

OptimizationRobotic Intelligence

🎯 What it does: Proposed and implemented a space-position-based visual servoing, obstacle-avoiding shape control framework for 11-degree-of-freedom hybrid continuum robots.

A Spatiotemporal Downwash Modeling for Agile Close-Proximity Multirotor Flight

Pavel Kharitenko, Yang Wang

Robotic IntelligenceTime SeriesPhysics Related

🎯 What it does: Studied the downwash interaction of a quadrotor in the high-speed range of 0.5-4.0 m/s through high-precision CFD simulations, and proposed a geometric deep neural network prediction model that simultaneously considers both absolute and relative velocities.

A stepwise identification framework for determining the physical feasibility parameters of robot dynamics

Guanghui Liu, Qiang Li

OptimizationRobotic IntelligencePhysics Related

🎯 What it does: Proposed a four-step systematic framework for identifying physically feasible robot dynamics parameters.

A Strawberry Harvesting Tool with Minimal Footprint

Mohamed Sorour, P. From

Agriculture Related

🎯 What it does: Proposed a prototype for strawberry picking in desktop cultivation with an extremely small footprint, using a smooth gripper to guide the stem into a precise groove and remote laser beams for cutting.

A Study on the Generation of Single Cell Droplets via the Combination of Lateral-Field Optoelectronic Tweezers and Electrowetting-on-Dielectric

Jiawei Zhao, Lin Feng

Biomedical Data

🎯 What it does: Combining Lateral-Field Optoelectronic Tweezers (LOET) with Electrowetting-on-Dielectric (EWOD) technology, individual cells are isolated, encapsulated into separate droplets, generating single-cell droplets and enabling control over their microenvironment.

A Surgical State Identifying Method based on BiLSTM with Vibration Processing for Improving Safety of Bone Milling System*

Jinyu Liu, Yu Dai

ClassificationRecurrent Neural NetworkTime SeriesBiomedical Data

🎯 What it does: Studied a vibration signal processing method based on BiLSTM for high-precision identification of bone milling surgery states. Vibration signals were collected through experiments on pig scapulae, and the network was trained to recognize milling depth and bone layer ratio.

A System for Multi-View Mapping of Dynamic Scenes Using Time-Synchronized UAVs

Aniket Gupta, Hanumant Singh

Robotic IntelligenceSimultaneous Localization and MappingImageBenchmark

🎯 What it does: Developed a multi-perspective synchronous acquisition system based on drones, and provided a corresponding unrestricted outdoor scene synchronous multi-perspective image dataset.