arXivSub Start free trial

ICRA 2024 Papers with AI Summaries

IEEE International Conference on Robotics and Automation · 1760 papers

1 kHz Behavior Tree for Self-adaptable Tactile Insertion

Yansong Wu, Sami Haddadin

Robotic IntelligenceTime Series

🎯 What it does: Studied an adaptive tactile insertion skill framework based on behavior trees, which estimates contact states using high-frequency tactile data and performs primitive switching to improve insertion task performance.

17-Point Algorithm Revisited: Toward a More Accurate Way

Chen Xie, Fenghua He

Pose Estimation

🎯 What it does: Improving the 17-point relative pose estimation algorithm

2D-3D Object Shape Alignment for Camera-Object Pose Compensation in Object-Visual SLAM

H. Lee, Chan Gook Park

Pose EstimationOptimizationSimultaneous Localization and MappingVideo

🎯 What it does: Propose a 6-degree-of-freedom (6-DOF) object pose compensation method based on robust optimization, which aligns the 2D segmentation of objects on the image plane with the 3D projection of objects through shape alignment, and constructs an invariant extended Kalman filter (EKF)-based object visual SLAM with the relative pose between the camera and object as the measurement model; during the alignment process, the pose is decomposed into translation and rotation, where translation is optimized linearly and rotation is optimized in parallel using multiple initial values.

3D Autocomplete: Enhancing UAV Teleoperation with AI in the Loop

Batool Ibrahim, Daniel C. Asmar

Robotic IntelligenceReinforcement Learning from Human Feedback

🎯 What it does: Proposed a 3D Autocomplete framework that predicts user intent via AI and automatically completes UAV remote control actions in 3D space, enhanced with a mixed reality interface to improve human-computer interaction

3D Force and Contact Estimation for a Soft-Bubble Visuotactile Sensor Using FEM

Jing-Chen Peng, Kris Hauser

Robotic IntelligenceImage

🎯 What it does: Proposed a model-based method that utilizes the RGBD camera and pressure sensor inside the soft foam tactile sensor to reconstruct dense contact forces; simultaneously, a linear plane stress finite element model requiring calibration of only three variables was constructed.

3D Navigation of a Magnetic Swimmer Using a 2D Ultrasonography Probe Manipulated by a Robotic Arm for Position Feedback

Premal Gorroochurn, Julien Leclerc

Robotic IntelligenceImageUltrasound

🎯 What it does: Developed a method for three-dimensional navigation control of a magnetic swimmer using a 2D ultrasound probe and a robotic arm, and verified its position measurement and closed-loop control in experiments.

3D Object Detection with VI-SLAM Point Clouds: The Impact of Object and Environment Characteristics on Model Performance

Lin Duan, Maria Gorlatova

Object DetectionSimultaneous Localization and MappingPoint CloudBenchmark

🎯 What it does: Created and released two datasets, VIP500 and VIP500-D, and used them to evaluate the performance of three leading 3D object detection models on VI-SLAM point clouds, analyzing the impact of object and environmental features on model performance.

3D-BBS: Global Localization for 3D Point Cloud Scan Matching Using Branch-and-Bound Algorithm

Koki Aoki, Junichi Meguro

Pose EstimationOptimizationComputational EfficiencySimultaneous Localization and MappingPoint Cloud

🎯 What it does: Proposes 3D-BBS, a fast and accurate 3D global localization method that extends the original branch-and-bound based 2D scan matching algorithm

3D-OAE: Occlusion Auto-Encoders for Self-Supervised Learning on Point Clouds

Junsheng Zhou, Zhizhong Han

Representation LearningTransformerAuto EncoderPoint Cloud

🎯 What it does: Propose a self-supervised learning framework called 3D-OAE based on occlusion completion, which learns point cloud representations by randomly occluding local blocks of point clouds and completing them on the remaining points.

3QFP: Efficient neural implicit surface reconstruction using Tri-Quadtrees and Fourier feature Positional encoding

Shuo Sun, Martin Magnusson

GenerationNeural Radiance FieldMesh

🎯 What it does: Proposes a neural implicit surface reconstruction method that utilizes a sparse tri-quadtree structure and Fourier feature position encoding.

6-DoF Closed-Loop Grasping with Reinforcement Learning

Sverre Herland, E. Misimi

SegmentationRobotic IntelligenceReinforcement LearningImagePoint Cloud

🎯 What it does: Developed a vision-based 6-DoF closed-loop grasping framework that trains in a simulated environment using deep reinforcement learning and directly deploys to the real world without requiring any real-world grasping examples or fine-tuning;

6-DoF Grasp Pose Evaluation and Optimization via Transfer Learning from NeRFs

Gergely S'oti, Björn Hein

Pose EstimationOptimizationRobotic IntelligenceNeural Radiance Field

🎯 What it does: Proposed and implemented a grasp evaluation model based on implicit behavioral cloning, trained with a small amount of demonstration data, and optimized 6-DoF grasp candidates using the implicit visual and geometric features from pre-trained NeRF.

A 3D Mixed Reality Interface for Human-Robot Teaming

Jiaqi Chen, Hermann Blum

Robotic IntelligenceSimultaneous Localization and Mapping

🎯 What it does: A mixed reality human-robot collaboration system is proposed, allowing operators to view the robot's location in real-time, visualize the 3D environment map constructed by the robot, and use drag-and-drop interaction to move the robot to a new target location.

A 3D Vector Field and Gaze Data Fusion Framework for Hand Motion Intention Prediction in Human-Robot Collaboration

Maleen Jayasuriya, Dikai Liu

ClassificationComputational EfficiencyRobotic IntelligenceMultimodalityTime Series

🎯 What it does: Propose a framework that combines hand trajectory and gaze data to achieve real-time prediction of hand motion intent with low computational resources

A Backdrivable Axisymmetric Kinematically Redundant (6+3)-Degree-of-Freedom Hybrid Parallel Manipulator

Jehyeok Kim, Clément Gosselin

Robotic Intelligence

🎯 What it does: Propose a hybrid parallel robot with six degrees of freedom plus three redundant degrees of freedom. By setting the first rotational joints of each leg to the same rotational axis, an axis-symmetric workspace is achieved. Kinematic redundancy is utilized to completely avoid Type II singularities, and a gripper mechanism is designed to further expand the orientation workspace. By controlling redundant degrees of freedom to maintain a constant angle, the orientation workspace is enhanced. The robot achieves a workspace-to-benchmark-area ratio comparable to serial robots and features back-drivable characteristics with direct drive or quasi-direct drive actuators.

A Bayesian Optimization Framework for the Automatic Tuning of MPC-based Shared Controllers

A. V. D. Horst, Elena Torta

OptimizationRobotic Intelligence

🎯 What it does: Proposed a Bayesian optimization framework for automatically tuning MPC-based shared controllers, applied to image-guided therapy robots.

A Biomorphic Whisker Sensor for Aerial Tactile Applications

Chaoxiang Ye, S. Hamaza

Robotic IntelligenceRecurrent Neural NetworkTime Series

🎯 What it does: Design, fabricate, and evaluate a lightweight follicle sensor based on MEMS barometers, and propose a recursive multi-output network (RMN) for 3D touch point localization.

A CBF-Adaptive Control Architecture for Visual Navigation for UAV in the Presence of Uncertainties

V. N. Sankaranarayanan, G. Nikolakopoulos

Autonomous DrivingRobotic IntelligenceImage

🎯 What it does: Propose a vision-based safe transmission control scheme for quadrotor UAVs, utilizing visual locking control barrier function (VCBF) and parameterized switched decreasing control barrier function (DCBF) to achieve positioning, landing, and visual locking between surface robots, ensuring safety under modeling uncertainties and external disturbances.

A Coarse-to-Fine Place Recognition Approach using Attention-guided Descriptors and Overlap Estimation

Chencan Fu, Yong Liu

RecognitionAutonomous DrivingSimultaneous Localization and MappingVideoPoint Cloud

🎯 What it does: Propose a coarse-to-fine place recognition method that integrates BEV feature extraction, coarse matching, and fine verification;

A Collision-Aware Cable Grasping Method in Cluttered Environment

Lei Zhang, Jianwei Zhang

Data SynthesisRobotic IntelligenceConvolutional Neural Network

🎯 What it does: Developed CG-CNN for cable grasping with collision awareness in cluttered environments, generating data via physics simulation and training through domain randomization.

A Combination of a Controllable Clutch and an Oscillating Slider Crank Mechanism for Ease of Direct-Teaching with Various Payloads

Muhammad Arifin, Shigeki Sugano

Robotic Intelligence

🎯 What it does: This paper proposes and verifies a variable load gravity compensation method using a series clutch drive to achieve high backdrivability during direct teaching processes.

A Compiler Framework for Proactive UAV Regulation Enforcement

Huaxin Tang, Yu David Liu

Safty and Privacy

🎯 What it does: Developed the Themis1 compiler framework, achieving automated, proactive drone regulation enforcement.

A Constrained Path Following Method for Snake-like Manipulators via Controlled Winding Uncoiling Strategy

Mingrui Luo, Min Tan

Robotic Intelligence

🎯 What it does: A constraint path following method for a snake-like robotic arm based on a controlled winding-unwinding strategy was developed.

A Control Barrier Function-based Motion Planning Scheme for a Quadruped Robot

Halil Utku Unlu, F. Khorrami

Robotic Intelligence

🎯 What it does: Designed a motion planning algorithm based on Control Barrier Functions (CBF) that can explore and reach the target point in unknown environments while providing velocity commands to the robot's control module.

A Data-Driven Approach to Geometric Modeling of Systems with Low-Bandwidth Actuator Dynamics

Siming Deng, N. Cowan

OptimizationRobotic Intelligence

🎯 What it does: Developed a data-driven modeling framework based on geometric mechanics (gauge theory) for soft robots with low-bandwidth actuators, constructing a coherent model that integrates actuators and kinematics, and validated it on simplified swimming robot models and stimulus-responsive hydrogel simulators; simultaneously proposed a method for numerically optimizing control signals through iterative model refinement to optimize the input waveform of hydrogel crawlers.

A Deep Learning Framework for Non-Symmetrical Coulomb Friction Identification of Robotic Manipulators

M. Lahoud, F. Cannella

Robotic IntelligencePhysics Related

🎯 What it does: Proposed and validated an enhanced DeLaN architecture for identifying linear viscous and asymmetric Coulomb friction, and learning parameters such as motor rotor inertia; validated on the real-world UR5e robot.

A Detachable FBG-Based Contact Force Sensor for Capturing Gripper-Vegetable Interactions

W. Lai, S. Phee

Agriculture Related

🎯 What it does: Developed a detachable flexible contact force sensor based on fiber Bragg grating (FBG) for capturing interactions during robotic grasping of vegetables

A Distributed Multi-Robot Framework for Exploration, Information Acquisition and Consensus

Aalok Patwardhan, Andrew J. Davison

OptimizationRobotic Intelligence

🎯 What it does: Construct a distributed multi-robot framework that uses factor graphs to represent tasks such as local planning, global goal coordination, and collaborative mapping, and employs Gaussian Belief Propagation (GBP) for information acquisition and decision-making.

A Distributed Processing Approach for Smooth Task Transitioning in Strict Hierarchical Control

Francesco Tassi, Arash Ajoudani

OptimizationComputational Efficiency

🎯 What it does: A distributed processing method is used to achieve smooth task switching under strict hierarchical control.

A Dragonfly-inspired Flapping Wing Robot Mimicking Force Vector Control Approach

Fangyuan Liu, Zhan Tu

OptimizationRobotic Intelligence

🎯 What it does: Designed, optimized, and manufactured a dragonfly-inspired flapping-wing robot (DFWR) with tiltable stroke plane control, and demonstrated its 'force vectorization' capability through theoretical analysis and experimental validation.

A Du-Octree based Cross-Attention Model for LiDAR Geometry Compression

Mingyue Cui, Kai Huang

CompressionTransformerPoint Cloud

🎯 What it does: Propose a dual octree structure and design a cross-attention model to capture hierarchical features between different octrees, with each octree employing a Transformer-based deep entropy model and an arithmetic encoder.

A Dual Closed-Loop Control Strategy for Human-Following Robots Respecting Social Space

Jianwei Peng, Houde Dai

Robotic Intelligence

🎯 What it does: Proposes a dual-loop human-following control strategy combining Model Predictive Control (MPC) and impedance control, enabling robots to follow targets while maintaining social space and avoiding obstacles.

A Flat Tendon-Driven Continuum Microrobot for Brain Interventions

L. Noseda, M. Sakar

Robotic IntelligenceBiomedical Data

🎯 What it does: Designed and fabricated a planar ribbon tension-driven continuum micro-robot specifically for brain tissue navigation, and conducted in vitro characteristic evaluation and testing on brain tissue models.

A Force-Controlled Gripper Capable of Measuring Mechanical Properties of an Object

Yi-Shian Tsai, Chao-Chieh Lan

Robotic Intelligence

🎯 What it does: Designed and implemented a dual-finger force-controlled gripper based on the deformation perception of elastic elements in the drive system; achieved precise position and force measurement at various gripper positions using the minimal number of optical encoders, and can accurately measure the dimensions and stiffness of objects during grasping.

A Force-driven and Vision-driven Hybrid Control Method of Autonomous Laparoscope-Holding Robot

Jin Fang, Yanwei Qu

Robotic IntelligenceBiomedical Data

🎯 What it does: A laparoscopic instrument-holding robotic system utilizing both force-driven and vision-driven dual mechanisms was developed, enabling continuous and large-range field of view (FOV) adjustment, and equipped with an integrated tactile handle for human-robot interaction.

A Framework for Real-time Generation of Multi-directional Traversability Maps in Unstructured Environments

Tao Huang, Shuxin Wang

Autonomous DrivingComputational EfficiencyKnowledge DistillationRepresentation LearningConvolutional Neural Network

🎯 What it does: Proposes a real-time multi-direction traversability map (MTraMap) generation framework for unstructured environments; first trains a single-direction traversability classifier UniTraT using self-supervised learning, then obtains a multi-direction traversability network MultiTCNN through Uni-to-multi-direction traversability distillation (UMTraDistill), which can directly generate MTraMap.

A Generalized Acquisition Function for Preference-based Reward Learning

Evan Ellis, Erdem Biyik

Reinforcement Learning from Human FeedbackReinforcement LearningText

🎯 What it does: Proposed a generic acquisition function for preference-based reward learning, aiming to learn behavioral equivalence classes of the reward function rather than complete parameters.

A Generic Modeling Framework For the Design of Tendon-Driven Continuum Manipulators with Flexure Patterns

Yang Liu, F. Alambeigi

Robotic Intelligence

🎯 What it does: Proposed a new mathematical framework for modeling the deformation behavior of TD-CM with discontinuous cross-sectional geometries (flexible modes), and introduced the concept of design space, enabling intuitive analysis of the deformation behavior space through geometric design parameters; subsequently, the performance of this framework was validated through simulation and experiments.

A GP-based Robust Motion Planning Framework for Agile Autonomous Robot Navigation and Recovery in Unknown Environments

Nicholas Mohammad, N. Bezzo

Robotic Intelligence

🎯 What it does: Propose a Gaussian Process (GP) model for actively predicting the risk of motion planning failure in mobile robots, and trigger recovery actions based on the same model when the risk exceeds a threshold, seeking safe states to proceed.

A Group Theoretic Metric for Robot State Estimation Leveraging Chebyshev Interpolation

Varun Agrawal, F. Dellaert

Robotic IntelligenceSimultaneous Localization and MappingPoint Cloud

🎯 What it does: Propose a new metric for robot state estimation based on the SE2(3) Lie group, calculate linear velocity from motion capture data using Chebyshev interpolation and pseudo-spectral parameterization methods, and validate the metric on multiple robot platforms while comparing it with the finite difference method.

A Guided Gaussian-Dirichlet Random Field for Scientist-in-the-Loop Inference in Underwater Robotics

Chad R. Samuelson, Joshua G. Mangelson

Robotic Intelligence

🎯 What it does: Propose Guided Gaussian-Dirichlet Random Field (GGDRF) and introduce NLP's Dirichlet Forest priors into visual topic modeling, achieving semi-supervised learning in the field of underwater robotics.

A Helical Bistable Soft Gripper Enable by Pneumatic Actuation

Xuanchun Yin, Jiantao Zhang

Robotic Intelligence

🎯 What it does: Proposed and implemented a bistable soft spiral gripper, embedding a ready-made bistable steel shell into a 3D-printed flexible exoskeleton, and achieving self-rolling and grasping without energy consumption through pneumatic actuation.

A Hierarchical Framework for Robot Safety using Whole-body Tactile Sensors

Shuo Jiang, Lawson L. S. Wong

Robotic Intelligence

🎯 What it does: Proposes a three-layer granularity safety framework based on full-body tactile sensors, which can execute safe maneuvers when detecting hazardous stimuli.

A Hybrid Admittance Control Algorithm for Automatic Robotic Cranium-Milling

Chen Qian, Guibin Bian

Robotic IntelligenceBiomedical Data

🎯 What it does: A hybrid impedance control algorithm was proposed for automated robotic craniotomy, combining model-independent adaptive nonlinear force control and fuzzy control, with its effectiveness validated through 42 cases of ex vivo animal craniotomy experiments.

A Hybrid Dynamical Model for Robotic Underwater Vehicles when Submerged or Surfaced: Approach and Preliminary Evaluation

J. Hunt, L. Whitcomb

Robotic IntelligencePhysics Related

🎯 What it does: Designed and evaluated a Monte Carlo-based hybrid model for calculating buoyancy and righting moment of partially or fully submerged underwater vehicles, thereby simulating their interaction with the free surface.

A Large-area Tactile Sensor for Distributed Force Sensing Using Highly Sensitive Piezoresistive Sponge

Wendong Zheng, Huaping Liu

Robotic Intelligence

🎯 What it does: Researchers used multi-walled carbon nanotubes (MWCNT) and polyurethane (PU) sponge to prepare a high-sensitivity piezoresistive sponge, and based on this, designed a flexible large-area tactile sensor using resistance imaging technology for distributed force sensing.

A Large-scale Suction-based Climbing Parallel Robot for Wall Painting Application

Abdur Rosyid, B. El-Khasawneh

Robotic Intelligence

🎯 What it does: This paper proposes a large-scale adsorption climbing robot that uses a three-degree-of-freedom parallel mechanism for movement and achieves wall operations through three sets of suction cups;

A Learning-Based Framework for Safe Human-Robot Collaboration with Multiple Backup Control Barrier Functions

Neil C. Janwani, J. Burdick

Autonomous DrivingOptimizationSafty and PrivacyRobotic IntelligenceRecurrent Neural Network

🎯 What it does: Propose a learning-based safety control framework that utilizes switching among multiple backup controllers, formally guaranteeing robot safety under constrained control inputs through backup control barrier functions (BCBF), while satisfying driver intent.

A Magnetic Continuum Robot with In-situ Magnetic Reprogramming Capability

Junnan Xue, Li Zhang

Robotic Intelligence

🎯 What it does: Developed a reprogrammable magnetic continuum robot (RMCR) capable of in-situ magnetic field reprogramming and demonstrated its complex deformation capabilities.

A Metacognitive Approach to Out-of-Distribution Detection for Segmentation

Meghna Gummadi, Eric Eaton

SegmentationAnomaly Detection

🎯 What it does: Proposed a lightweight meta-cognitive module that real-time detects pixel-level out-of-distribution (OOD) data through entropy values, segmentation prediction, and spatial context.

A Meter-scale Ornithopter Capable of Jumping Take-off

Wei Yan, Hao Wang

Robotic Intelligence

🎯 What it does: Achieved the function of takeoff through jumping for a meter-scale flapping-wing UAV on flat ground, eliminating traditional mechanical legs and adopting a three-legged structure consisting of two wings and three carbon fiber springs;

A Method for Multi-Robot Asynchronous Trajectory Execution in MoveIt2

Pascal Stoop, G. T. Carughi

Robotic Intelligence

🎯 What it does: Propose a method for multi-robot/multi-arm systems in MoveIt2 to execute trajectories in parallel independently within a shared workspace.

A Miniature 1R1T Precision Manipulator with Remote Center of Motion for Minimally Invasive Surgery

Hiroyuki Suzuki

Robotic Intelligence

🎯 What it does: Designed and fabricated a miniature 1R1T precision manipulator ORIGANOID with dimensions of 60×120×30 mm and a weight of 12.6 g.

A Model for Multi-Agent Autonomy That Uses Opinion Dynamics and Multi-Objective Behavior Optimization

Tyler M. Paine, Michael R. Benjamin

Robotic Intelligence

🎯 What it does: A hierarchical architecture is proposed, using a nonlinear dynamic opinion process to model high-level group selection and multi-objective behavior optimization to model individual decision-making, in order to control the behavior of multi-robot systems.

A Modular Aerial System Based on Homogeneous Quadrotors with Fault-Tolerant Control

Mengguang Li, Heinz Koeppl

Robotic IntelligencePhysics Related

🎯 What it does: Proposed the modular drone system IdentiQuad, composed of homogeneous quadrotor modules, where individual units can fly like traditional quadcopters, or enhance control degrees of freedom through task-specific customization; a generic controller was designed to tolerate rotor failures and balance energy consumption across modules.

A Modular, Tendon Driven Variable Stiffness Manipulator with Internal Routing for Improved Stability and Increased Payload Capacity

Kyle L. Walker, Francesco Giorgio-Serchi

Robotic IntelligencePhysics Related

🎯 What it does: Designed and experimentally validated a modular, tendon-driven bead-based continuum manipulator with adjustable stiffness and internal wire arrangement to enhance stability and load capacity.

A Movable Microfluidic Chip with Gap Effect for Manipulation of Oocytes

Shuzhang Liang, Fumihito Arai

Robotic IntelligenceBiomedical Data

🎯 What it does: Proposed a movable microfluidic chip integrated with a robotic manipulator for the precise release and manipulation of individual oocytes.

A Multi-modal Hybrid Robot with Enhanced Traversal Performance*

Zhipeng He, Hongbin Deng

Robotic Intelligence

🎯 What it does: Designed and implemented a multimodal hybrid robot with variable wheel diameter and foldable structure, enabling switching between flight and wheeled modes and enhancing obstacle crossing performance.

A Multi-Stable Curved Line Shape Display

W. A. Law, Sean Follmer

OptimizationRobotic IntelligencePhysics RelatedOrdinary Differential Equation

🎯 What it does: A multistable curve shape display device was constructed, utilizing circular splines to initialize discrete elastic rod simulations, generating drive commands to achieve 16 shapes and 8 boundary conditions of variable surfaces.

A Multifidelity Sim-to-Real Pipeline for Verifiable and Compositional Reinforcement Learning

Cyrus Neary, U. Topcu

Robotic IntelligenceReinforcement Learning

🎯 What it does: Proposed and demonstrated a combinatorial framework for training and validating reinforcement learning (RL) systems in a multi-fidelity simulation-to-real (sim-to-real) pipeline, which can decompose complex robotic tasks into subtasks and define their mathematical interfaces, enabling independent training and testing of subtask policies while ensuring the overall behavior after combination; the framework was experimentally applied to successfully train and deploy the manipulation of the Warthog unmanned ground robot.

A Multimodal Handover Failure Detection Dataset and Baselines

Santosh Thoduka, P. Plöger

Anomaly DetectionRobotic IntelligenceConvolutional Neural NetworkVideoMultimodalityBenchmark

🎯 What it does: Proposes a multimodal handoff failure detection dataset and presents two baseline methods.

A Neural-Evolutionary Algorithm for Autonomous Transit Network Design

Andrew Holliday, Gregory Dudek

Autonomous DrivingOptimizationGraph Neural NetworkBenchmark

🎯 What it does: Propose a hybrid algorithm combining graph neural network strategies and evolutionary algorithms for route planning in autonomous bus networks.

A Neuromorphic System for the Real-time Classification of Natural Textures

George Brayshaw, Martin J. Pearson

ClassificationComputational EfficiencySpiking Neural NetworkImage

🎯 What it does: Designed and implemented a real-time texture classification system using neuromorphic tactile sensors, synapse neural networks, and a novel decision algorithm.

A New Perspective of Deep Learning Testing Framework: Human-Computer Interaction Based Neural Network Testing

Wei Kong, Xiaohui Kuang

Safty and PrivacyAdversarial Attack

🎯 What it does: Proposed a deep learning security testing framework based on human-computer interaction, aiming to design more effective test cases and evaluate their quality

A non-cubic space-filling modular robot

Tyler Hummer, S. Kriegman

Robotic Intelligence

🎯 What it does: Fabricate 48 rhombic dodecahedron modular robots and assemble them into different superstructures, demonstrating the mobility of some of these structures.

A non-magnetic dual-mode linear pneumatic actuator: initial design and assessment

T. Portha, Pierre Renaud

Biomedical DataMagnetic Resonance Imaging

🎯 What it does: Designed and evaluated a dual-mode magnetic-free linear pneumatic actuator suitable for harsh environments such as MRI, capable of step/continuous motion modes and disengagement to allow direct operator control.

A Nonlinear Estimator for Dead Reckoning of Aquatic Surface Vehicles Using an IMU and a Doppler Velocity Log

Jessica Paterson, K. Groves

Robotic IntelligenceTime Series

🎯 What it does: Propose a method utilizing an adaptive nonlinear estimator to fuse Doppler Velocity Log (DVL) and Inertial Measurement Unit (IMU) for landmark-free localization of surface robots.

A novel algorithmic approach to obtaining maneuverable control-invariant sets *

Prashant Solanki, Coen C. de Visser

Optimization

🎯 What it does: Developed an algorithm to obtain control-invariant sets (maneuver set) that enable reaching any other state from any state within a given time frame without leaving the set, and verified its control invariance and reachability through mathematical proof.

A Novel Benchmarking Paradigm and a Scale- and Motion-Aware Model for Egocentric Pedestrian Trajectory Prediction

Amir Rasouli

Autonomous DrivingMultimodalityBenchmark

🎯 What it does: Proposes a new context-based evaluation paradigm and a new multimodal information fusion-based pedestrian trajectory prediction model with a positive perspective

A Novel Compact Design of a Lever-Cam based Variable Stiffness Actuator: LC-VSA

Hongxi Zhu, Ulrike Thomas

Robotic Intelligence

🎯 What it does: Proposed a new compact mechanical lever-cam structure variable stiffness actuator design

A Novel Fish-inspired Self-adaptive Approach to Collective Escape of Swarm Robots Based on Neurodynamic Models

Junfei Li, Simon X. Yang

Robotic Intelligence

🎯 What it does: The study proposes a fish-inspired heuristic adaptive method based on neural dynamics, achieving coordinated escape of swarm robots from threats.

A Novel Funnel-Based L1 Adaptive Fuzzy Approach for the Control Of An Actuated Ankle Foot Orthosis

Oussama Bey, S. Mohammed

🎯 What it does: Proposed a funnel-based adaptive L1 fuzzy control strategy for assisting ankle motion in powered ankle-foot orthoses (AAFOs).

A novel metric for detecting quadrotor loss-of-control

Jasper J. van Beers, Coen C. de Visser

Anomaly Detection

🎯 What it does: Proposed and verified a new metric based on actuator capability for detecting loss of control in quadrotor drones

A Novel Model for Layer Jamming-based Continuum Robots

Bowen Yi, Dikai Liu

Robotic Intelligence

🎯 What it does: Proposed and verified a control-oriented dynamic model for variable stiffness in Layer Jamming continuum robots, and conducted theoretical analysis and experimental validation of its shape locking and adjustable stiffness characteristics.

A Novel Robotic Bronchoscope with a Spring-Based Extensible Segment for Improving Steering Ability

Jie Wang, Hongen Liao

Robotic Intelligence

🎯 What it does: Developed a 4 mm diameter robotic bronchoscope with a spring-actuated collapsible section, controlled by two drive rods to achieve follow-the-leader or fixed-angle motion patterns. It can shorten to increase stiffness upon reaching the target location, providing a stable surgical platform.

A Novel SEA-based Haptic Interface for Robot-Assisted Vascular Interventional Surgery

Yonggang Yan, Jianmin Liu

Safty and PrivacyRobotic IntelligenceBiomedical Data

🎯 What it does: Designed and implemented a tactile interface capable of simultaneously capturing translational displacement, torsion angle, and gripping force information from the operator, generating real-time force feedback through a bidirectional force feedback control strategy, and supporting remote grip state control and safe operation force threshold settings.

A Novel Wide-Area Multiobject Detection System with High-Probability Region Searching

Xianlei Long, Qingyi Gu

Object DetectionImage

🎯 What it does: Propose a hybrid visual system integrating wide-angle cameras, high-speed search cameras, and electro-optical mirrors to achieve wide-area multi-target detection

A Novel, Efficient and Accurate Method for Lidar Camera Calibration

Zhanhong Huang, Xinming Huang

Autonomous DrivingImagePoint Cloud

🎯 What it does: Propose a new method for LiDAR and camera calibration using a uniquely designed acrylic chessboard

A Passive Variable Impedance Control Strategy with Viscoelastic Parameters Estimation of Soft Tissues for Safe Ultrasonography

Luca Beber, L. Palopoli

OptimizationSafty and PrivacyBiomedical DataUltrasound

🎯 What it does: Developed a safe ultrasound imaging method based on passive variable damping control strategy combined with soft tissue viscoelastic parameter estimation

A Passively Bendable, Compliant Tactile Palm with RObotic Modular Endoskeleton Optical (ROMEO) Fingers

Sandra Q. Liu, Edward H. Adelson

Robotic Intelligence

🎯 What it does: Developed a passively bendable, structurally flexible soft palm and designed the ROMEO two-jointed soft fingers capable of accommodating a low-cost flexible lighting system and high-resolution tactile sensors (inspired by GelSight). These components were subsequently integrated into different palm configurations, ultimately resulting in a soft robotic hand slightly larger than a baseball, featuring actuated fingers, a passive flexible palm, and full-hand high-resolution tactile perception.

A Personalizable Controller for the Walking Assistive omNi-Directional Exo-Robot (WANDER)

A. Fortuna, Arash Ajoudani

OptimizationRobotic Intelligence

🎯 What it does: Designed and evaluated a personalized controllable omnidirectional exoskeleton WANDER, which captures user intent through force/torque sensors, employs variable compliance control, and utilizes Preference-Based Optimization (PBO) to select compliance parameters for adaptation to different users;

A Phase-Change Emulsion Jamming Gripper for Manipulation of Micro-Scale Textured Surfaces

Alex Keller, Jonathan Rossiter

Robotic Intelligence

🎯 What it does: A new method utilizing shape-memory silicone sponge for adaptive grasping on micro-scale textured surfaces is proposed.

A Planar Compliant Contact Control Applied to Multi-dimensional Elastic Gripper for Unexpected Contact

Junnan Huang, Bin Liang

Robotic Intelligence

🎯 What it does: Propose a planar compliant contact control method, utilizing thin film gel and a two-finger gripper to design a multi-dimensional elastic gripper. An elastic interaction controller is designed through modeling to actively adjust gel deformation in planar operations, achieving desired contact forces and torques while avoiding obstacles.

A Point-to-distribution Degeneracy Detection Factor for LiDAR SLAM using Local Geometric Models

Sehua Ji, Haifei Zhu

Autonomous DrivingSimultaneous Localization and MappingPoint Cloud

🎯 What it does: A LiDAR-SLAM degradation detection method based on a local geometric model using point-to-distribution matching is proposed, aiming to address the matching degradation caused by insufficient geometric features in environments such as corridors and tunnels.

A Powerline Inspection UAV Equipped with Dexterous, Lockable Gripping Mechanisms for Autonomous Perching and Contact Rolling

Angus Lynch, Minas Liarokapis

Robotic IntelligenceSimultaneous Localization and MappingImage

🎯 What it does: Studied a quadrotor drone equipped with a scrollable stop mechanism and a depth vision system, achieving autonomous stopping and rolling inspection along power lines.

A probabilistic approach for learning and adapting shared control skills with the human in the loop

Gabriel Quere, João Silvério

Robotic IntelligenceSequential

🎯 What it does: Propose a method for learning active constraints from demonstrated end-effector trajectories using shared control templates, and demonstrate its successful application on the assistive robot EDAN for object grasping and adaptation to new environmental constraints.

A Probabilistic Motion Model for Skid-Steer Wheeled Mobile Robot Navigation on Off-Road Terrains

Ananya Trivedi, T. Padır

OptimizationRobotic IntelligenceBenchmark

🎯 What it does: Utilize Gaussian Process Regression and Sigma-point transformation to estimate the nonlinear effects of tire-terrain interaction on SSWMR speed, and propose a data-driven dynamic motion model to improve future pose prediction.

A Reinforcement Learning-based Control Strategy for Robust Interaction of Robotic Systems with Uncertain Environments

Diletta Sacerdoti, Cristian Secchi

Robotic IntelligenceReinforcement Learning

🎯 What it does: Propose a collaborative control strategy based on passive control and deep reinforcement learning, utilizing an energy tank to ensure stability, while online learning strategies estimate task demands and adapt control.

A Retinex Structure-based Low-light Enhancement Model Guided by Spatial Consistency

Miao Zhang, Shuai Lu

RestorationImage

🎯 What it does: Proposes SCRNet, a low-light image enhancement model based on the Retinex structure and guided by the principle of spatial consistency, which integrates channel, semantic, and texture consistency levels to adaptively enhance image features.

A Robust Model Predictive Controller for Tactile Servoing

Shuai Wang, Qiang Li

OptimizationRobotic Intelligence

🎯 What it does: Propose a data-driven model predictive controller (DDMPC) that calculates motion commands based on previous interaction experiences and feature deviations in the tactile space.

A Safety-Adapted Loss for Pedestrian Detection in Autonomous Driving

Maria Lyssenko, Rudolph Triebel

Object DetectionAutonomous DrivingConvolutional Neural NetworkPoint Cloud

🎯 What it does: Propose a safety-adapted loss function for training pedestrian detection models in autonomous driving scenarios, incorporating criticality information of each pedestrian to improve the training process.

A scalable monolithic 3D printable variable stiffness mechanism

Paul Baisamy, F. G. Serchi

🎯 What it does: Proposed a variable stiffness mechanism C-MAC that can be manufactured via single-piece 3D printing, and tested six configurations of it

A scalable, light-controlled, individually addressable, non-metal actuator array

Sophie Paul, E. Hawkes

Robotic Intelligence

🎯 What it does: Designed and demonstrated a scalable, light-controlled, individually addressable non-metallic helical crawling actuator array for achieving precise antenna rotation control.

A Semi-Autonomous Data-Driven Shared Control Framework for Robotic Manipulation and Cutting of an Unknown Deformable Tissue

Nicholas A. Strohmeyer, F. Alambeigi

Robotic IntelligenceBiomedical Data

🎯 What it does: Developed and demonstrated a semi-autonomous shared control framework for sharing unknown deformable tissue (U-DT) manipulation and cutting tasks between remote surgeons and surgical robots, achieving autonomous tension regulation and remote cutting via the da Vinci Research Kit (dVRK) platform;

A Simple Computationally Efficient Path ILC for Industrial Robotic Manipulators

M. Schwegel, A. Kugi

Computational EfficiencyRobotic Intelligence

🎯 What it does: Designed and verified a numerically efficient, sustainably executable scheme combining path iterative learning control (ILC) with model-based control to enhance the absolute precision of industrial robots.

A Soft Micro-Robotic Catheter for Aneurysm Treatment: A Novel Design and Enhanced Euler-Bernoulli Model with Cross-Section Optimization

Emanuele Nicotra, Thanh Nho Do

OptimizationRobotic IntelligenceOrdinary Differential Equation

🎯 What it does: Designed and verified a soft microrobotic catheter with a hollow channel, only 0.8 mm in diameter, capable of actively controlling the bending tip to achieve precise navigation to cerebral aneurysm targets and deliver embolic materials.

A soft miniaturized continuum robot with 3D shape sensing via functionalized soft optical waveguides

Viola Del Bono, S. Russo

Pose EstimationRobotic Intelligence

🎯 What it does: A fully soft, small-scale continuous robot was developed, integrating three-dimensional optical shape sensing achieved through functionalized soft optical waveguides.

A Soft Robot Inverse Kinematics for Virtual Reality

James M. Bern, Josie Hughes

Robotic IntelligenceMesh

🎯 What it does: Propose a pipeline named VR-Soft IK to intuitively control the morphology (configuration) of any 3D soft robot through virtual reality (VR) environments leveraging computer graphics technology

A Study of Force-Free Control Framework for Industrial Manipulator Tasks Based on High-Pass Filter

Guanwei He, Beichen Ding

Robotic Intelligence

🎯 What it does: Designed and verified a force-free control (FFC) framework for industrial robotic arms based on a six-axis force/torque sensor, and suppressed mechanical resonance by integrating a high-pass filter on the sensor;

A Surprisingly Efficient Representation for Multi-Finger Grasping

Hengxu Yan, Cewu Lu

Computational EfficiencyRepresentation LearningRobotic Intelligence

🎯 What it does: This paper proposes an efficient representation tailored for multi-finger grasping, and builds a compact decision model based on this representation to generate quality scores for different grasping poses under conditions with only a few hundred to a few thousand samples.