IROS 2023 Papers with AI Summaries
IEEE/RSJ International Conference on Intelligent Robots and Systems · 1195 papers
→ IROS 2023 papers with code (94)
Each paper below shows an AI-generated one-line summary. Get the full 6-part summary (innovation, method, data, results, limitations) and search all 1195 IROS 2023 papers by keyword, author or institution —
free trial on arXivSub.
2-DOF Robot Arm with Variable Torque Limiters Realized by Electrostatic Film Motors
Masahiko Osada, Akio Yamamoto
Robotic Intelligence
🎯 What it does: The study uses an electrostatic actuator-driven 2-DOF robotic arm and designs a variable torque limiter
360FusionNeRF: Panoramic Neural Radiance Fields with Joint Guidance
Shreyas Kulkarni, S. Scherer
GenerationVision Language ModelNeural Radiance FieldImage
🎯 What it does: A semi-supervised learning framework based on NeRF, which generates new view renderings from a single 360° panorama using geometric supervision and semantic consistency guidance;
3D Laser-and-Tissue Agnostic Data-Driven Method for Robotic Laser Surgical Planning
Guangshen Ma, Patrick J. Codd
Robotic IntelligenceImageBiomedical Data
🎯 What it does: A fully data-driven 3D laser carbonization pit prediction model is proposed and applied to robotic laser surgery planning, aiming to generate carbonization pits aligned with the target surface to minimize damage to healthy tissue;
3D Skeletonization of Complex Grapevines for Robotic Pruning
Eric Schneider, G. Kantor
Robotic IntelligenceAgriculture Related
🎯 What it does: A pipeline for complex grapevine 3D skeletonization was studied to enhance the perception capabilities required for robotic pruning.
3D-Aware Object Localization using Gaussian Implicit Occupancy Function
Vincent Gaudillière, D. Aouada
Object DetectionPose EstimationGaussian SplattingPoint CloudPhysics Related
🎯 What it does: Propose a 3D-aware target localization method based on an implicit Gaussian occupancy function, replacing traditional elliptical geometric parameters with Gaussian distribution parameters for 2D target regression.
6D Object Pose Estimation from Approximate 3D Models for Orbital Robotics
Maximilian Ulmer, Rudolph Triebel
Pose EstimationImage
🎯 What it does: Proposes a technique to estimate object 6D pose from a single image using only an approximate 3D model.
A Bayesian Reinforcement Learning Method for Periodic Robotic Control Under Significant Uncertainty
Yuanyuan Jia, T. Taniguchi
Robotic IntelligenceReinforcement Learning
🎯 What it does: Propose a three-stage periodic Bayesian reinforcement learning method for periodic robot control in uncertain environments.
A Bio-Inspired Robotic Finger: Mechanics and Control
Worathris Chungsangsatiporn, R. Chancharoen
Robotic Intelligence
🎯 What it does: Designed, manufactured, analyzed, and tested a biomorphic robotic finger.
A Bioinspired Underactuated Dual Tendon-Based Adaptive Gripper for Space Applications
Hamid Isakhani, Helya Isakhani
Robotic Intelligence
🎯 What it does: Designed and manufactured a bionic low-power mechanical hand with dual tendon drive and adaptive grasping capabilities, specifically tailored for aerospace applications;
A Cartesian Platform for Cooperative Multi-Robot Manipulation Tasks
S. Müller, Josie Hughes
OptimizationRobotic Intelligence
🎯 What it does: Designed and implemented a dual-arm collaborative robot system based on the Cartesian platform, enabling two sets of mirror robot arms to freely interact and complete collaborative manipulation tasks under 5 degrees of freedom.
A Co-Simulation Framework for Communication and Control in Autonomous Multi-Robot Systems
Srikrishna Acharya, Bharadwaj Amrutur
Robotic IntelligenceWorld Model
🎯 What it does: Proposed the CORNET 2.0 collaborative simulation framework, which can simultaneously and accurately simulate the behavior of multi-robot systems in physical environments as well as the communication network models within those environments.
A Comparison Between Framed-Based and Event-Based Cameras for Flapping-Wing Robot Perception
R. Tapia, A. Ollero
Robotic IntelligenceImage
🎯 What it does: This study experimentally compares the perception performance of frame cameras and event cameras on flapping-wing robots, and evaluates the performance of common vision algorithms on onboard data.
A Complementarity-Based Switch-Fuse System for Improved Visual Place Recognition
Maria Waheed, Shoaib Ehsan
RecognitionImage
🎯 What it does: Proposed and implemented a system called Switch-Fuse, which combines switching and fusing techniques in visual place recognition (VPR), first selecting the most suitable VPR technique for fusion based on the query image.
A Composite Control Strategy for Quadruped Robot by Integrating Reinforcement Learning and Model-Based Control
Shangke Lyu, Donglin Wang
Robotic IntelligenceReinforcement Learning
🎯 What it does: Integrate deep reinforcement learning (DRL) with model-based control to develop a composite control strategy for quadruped robots, and verify its robustness and compliant contact capability against external disturbances through physical experiments.
A Data-Driven Approach to Synthesizing Dynamics-Aware Trajectories for Underactuated Robotic Systems
Anusha Srikanthan, N. Matni
OptimizationRobotic Intelligence
🎯 What it does: Propose a data-driven dynamics-aware trajectory generation method.
A Decision Tree-based Monitoring and Recovery Framework for Autonomous Robots with Decision Uncertainties
Rahul Peddi, N. Bezzo
Anomaly DetectionRobotic Intelligence
🎯 What it does: Proposes a monitoring and recovery framework based on decision trees, applied to fault detection and recovery in autonomous robots under decision uncertainty.
A Distributed Scheduling Method for Networked UAV Swarm based on Computing for Communication
Runfeng Chen, Yuchong Huang
Optimization
🎯 What it does: A distributed UAV swarm scheduling method based on computation-driven communication is proposed, which reduces communication volume and improves scheduling efficiency through new task removal strategies, task-related optimization, and historical bid information inference.
A Dynamic Programming Algorithm for Grid-Based Formation Planning of Multiple Vehicles
T. Au
OptimizationRobotic Intelligence
🎯 What it does: Propose a grid-based planning method that uses dynamic programming to compute approximate optimal motion plans for multiple vehicles from start to destination, aiming to minimize total completion time while ensuring safety.
A Fabrication and Simulation Recipe for Untethering Soft-Rigid Robots with Cable-Driven Stiffness Modulation
James M. Bern, Daniela Rus
Robotic Intelligence
🎯 What it does: A method is proposed to achieve significant and rapid stiffness variation through the compression and relaxation of cable-driven flexible components, and a differentiable modeling framework is developed for motion planning, integrating adjustable stiffness joints, tendons, rigid joints, and fundamental fluid dynamics. This method was used to create the cable-free soft rigid turtle robot SoRiTu, capable of performing various swimming motions.
A Fast and Map-Free Model for Trajectory Prediction in Traffics
Junhong Xiang, Zhixiong Nan
Autonomous DrivingRecurrent Neural NetworkGraph Neural NetworkTransformerSequential
🎯 What it does: Proposed an efficient trajectory prediction model that does not rely on traffic maps.
A Fast Steerable Soft Robot for Navigating a Pipe Network
Ian Stewart, Phanindra Tallapragada
Robotic Intelligence
🎯 What it does: Developed a two-unit rotatable vibratory bristle robot, and demonstrated through modeling and experiments its full-range control in pipeline networks, including moving forward, turning left/right, and ascending/descending vertical pipes.
A Finite-Time State-Dependent Differential Riccati Equation Control Design for Closed-Loop SMA-Actuated Hip Joint
V. Perez-Sanchez, A. Ollero
OptimizationRobotic IntelligencePhysics RelatedOrdinary Differential Equation
🎯 What it does: Modeling and designing closed-loop control for a winged flying robot (SMA-driven hip joint).
A Framework for Few-Shot Policy Transfer Through Observation Mapping and Behavior Cloning
Yash Shukla, Jivko Sinapov
Domain AdaptationReinforcement LearningGenerative Adversarial Network
🎯 What it does: Proposed a framework that achieves few-shot policy transfer through observation mapping and behavior cloning.
A Game-Theoretic Framework for Joint Forecasting and Planning
K. Kedia, Sanjiban Choudhury
Robotic IntelligenceTime SeriesSequential
🎯 What it does: Propose a game theory framework for joint prediction and planning of robots in human environments, and provide a practical end-to-end trained algorithm.
A Gaussian Process Model for Opponent Prediction in Autonomous Racing
F. Busch, F. Borrelli
Autonomous DrivingOptimization
🎯 What it does: Built and validated a one-step opponent prediction model based on Gaussian processes, integrated with model predictive control (MPC) for safety-performance trade-off in autonomous racing.
A Generic Framework for Byzantine-Tolerant Consensus Achievement in Robot Swarms
Hanqing Zhao, Marco Dorigo
Robotic Intelligence
🎯 What it does: Proposes a generic framework based on blockchain smart contracts, enabling robot swarms to achieve secure consensus within any observation space and to identify and neutralize Byzantine robots.
A GM-PHD Filter with Estimation of Probability of Detection and Survival for Individual Targets
R. T. Perera, P. Stegagno
Object TrackingPoint Cloud
🎯 What it does: Proposed an improved GM-PHD filter that can estimate the detection probability and survival probability of each target online, eliminating the need for pre-set fixed values.
A Grasp Pose is All You Need: Learning Multi-Fingered Grasping with Deep Reinforcement Learning from Vision and Touch
Federico Ceola, Lorenzo Natale
Robotic IntelligenceReinforcement LearningVideo
🎯 What it does: Proposed and implemented a deep reinforcement learning method called G-PAYN for multi-fingered grasping tasks by the iCub humanoid hand in a simulated environment; the method initializes the policy by automatically collecting task demonstrations and combines grasp poses generated by external algorithms as the starting point of the grasping process; subsequently, the trained control policy completes the grasping; experiments were conducted on the MuJoCo simulator, using objects from the YCB-Video dataset for evaluation.
A Handle Robot for Providing Bodily Support to Elderly Persons
Jr Roberto Bolli, F. I. H. Harry Asada
OptimizationRobotic IntelligenceVideo
🎯 What it does: Developed a mobile robot that can provide a handle at any position in space to assist the elderly with daily mobility and provide physical support.
A Highly Maneuverable Flying Squirrel Drone with Controllable Foldable Wings
Jun-Gill Kang, Soohee Han
Robotic IntelligenceReinforcement Learning
🎯 What it does: Developed a foldable-wing drone inspired by flying squirrels, achieving higher maneuverability and stable tracking performance.
A Hybrid FNS Generator for Human Trunk Posture Control with Incomplete Knowledge of Neuromusculoskeletal Dynamics
Xuefeng Bao, M. Audu
Biomedical Data
🎯 What it does: Proposed a functional neuromuscular stimulation generator that combines a robust nonlinear controller with an artificial neural network (ANN) to control the trunk posture of spinal cord injury patients without complete knowledge of neuromuscular dynamics.
A Hybrid-State Path Planner for ASV Formations with Full-Scale Experiments
E. Ruud, Glenn Bitar
Autonomous DrivingOptimization
🎯 What it does: A hybrid state path planner for autonomous surface vessels (ASVs) subject to minimum turning radius constraints was developed, capable of finding the shortest path in environments with existing maps, operator-defined no-go zones, and static obstacles detected by the vehicle.
A Light-Weight, Low-Cost, and Sustainable Planning System for UAVs Using a Local Map Origin Update Approach
Dasol Lee, Sanghyun Joo
Computational EfficiencyRobotic IntelligenceSimultaneous Localization and MappingPoint Cloud
🎯 What it does: Proposed a sustainable planning system for small unmanned aerial vehicles (UAVs), achieving continuous planning through local map origin updates.
A Localization Framework for Boundary Constrained Soft Robots
Koki Tanaka, Matthew Spenko
Robotic IntelligenceSimultaneous Localization and Mapping
🎯 What it does: Proposed a localization framework for soft robots with boundary constraints (with special hardware structures), utilizing flexible boundaries to construct onboard sensors that measure the relative distance between sub-robots, and inputting the measurement data into a linear Kalman filter for precise localization.
A Mangasarian-Soldov Function Based Neural Network for Constrained Control of Parallel and Serial Robots
Weibing Li, Yongping Pan
OptimizationRobotic Intelligence
🎯 What it does: Proposed a zeroing neural network (MS-ZNN) solver based on the Mangasarian-Solodov function for solving kinematic control of parallel and serial robots under joint constraints.
A Minimal Collision Strategy of Synergy Between Pushing and Grasping for Large Clusters of Objects
Chong Chen, Quang Dan Le
Robotic Intelligence
🎯 What it does: Proposes a collaborative strategy utilizing push and grasp actions to automatically push, pull, and grasp large clusters of tightly stacked objects.
A Model Predictive Path Integral Method for Fast, Proactive, and Uncertainty-Aware UAV Planning in Cluttered Environments
Jacob Higgins, N. Bezzo
Autonomous DrivingOptimization
🎯 What it does: Proposes a recursive horizon motion planning method based on model predictive path integral (MPPI) control theory for fast, proactive, and uncertainty-sensitive UAV trajectory generation in crowded environments.
A Multiplicative Value Function for Safe and Efficient Reinforcement Learning
Nick Bührer, L. Gool
Reinforcement LearningImagePoint Cloud
🎯 What it does: Propose a safe model-free reinforcement learning algorithm that employs a multiplicative value function to separate the safety estimator from the reward estimator;
A Multitask and Kernel Approach for Learning to Push Objects with a Target-Parameterized Deep Q-Network
Marco Ewerton, J. Odobez
Robotic IntelligenceReinforcement Learning
🎯 What it does: Proposed a multi-head target parameterized DQN for learning strategies in robot pushing tasks.
A New 1-mg Fast Unimorph SMA-Based Actuator for Microrobotics
Conor K. Trygstad, N. O. Pérez-Arancibia
Robotic Intelligence
🎯 What it does: Proposed and fabricated a 1mg SMA-based miniature single curved surface actuator, using it to drive the MiniBug crawling robot and the WaterStrider water surface tension crawling robot
A New Design of Multilayered String Jamming Mechanism with Three-Degree-of-Freedom
Ryohei Michikawa, Fumitoshi Matsuno
Physics Related
🎯 What it does: A new design of a multi-layer string damping mechanism is proposed, and its performance in both flexible and rigid states is evaluated and experimentally validated.
A New Method Combining Single-Point Push and Double-Point Complete Push for Partially Observable Scenarios
Jiawei Zhang, Jifeng Guo
Robotic IntelligenceRecurrent Neural Network
🎯 What it does: Proposed an SDP-RRT algorithm for push path planning in partially observable environments with obstacles, combining single-point pushing and two-point full pushing.
A Novel Approximation for the Spring Loaded Inverted Pendulum Model of Locomotion
Ali U. Kilic, David J. Braun
OptimizationRobotic IntelligencePhysics RelatedOrdinary Differential Equation
🎯 What it does: Proposed a new approximation method for the Spring-Loaded Inverted Pendulum (SLIP) model, and designed a controller based on this approximation to achieve short-step stability;
A Novel Coiled Cable-Conduit-Driven Hyper-Redundant Manipulator for Remote Operating in Narrow Spaces
Mingrui Luo, Min Tan
Robotic Intelligence
🎯 What it does: Propose a ultra-redundant manipulator (C-CDHRM) driven by coiled cable conduits, and design a multi-layer control system to achieve precise remote operation in confined spaces.
A Novel Control Law for Multi-Joint Human-Robot Interaction Tasks While Maintaining Postural Coordination
Keya Ghonasgi, A. Deshpande
Robotic Intelligence
🎯 What it does: Developed a PCA-based control algorithm that achieves joint angle coordination control during multi-joint human-robot interaction by applying force fields, implemented on the Harmony exoskeleton; the algorithm can adjust damping stiffness in unstructured interactions to assist users in completing tasks at different speeds;
A Novel Lockable Spring-Loaded Prismatic Spine to Support Agile Quadrupedal Locomotion
Keran Ye, Konstantinos Karydis
Robotic IntelligenceBenchmark
🎯 What it does: Systematically studied the impact of flexible ratcheting columns on quadruped robot locomotion, developed a lockable spring-loaded ratcheting module and a new SCIQ platform.
A Novel Obstacle-Avoidance Solution With Non-Iterative Neural Controller for Joint-Constrained Redundant Manipulators
Weibing Li, Yongping Pan
OptimizationRobotic Intelligence
🎯 What it does: Proposed a non-iterative neural controller NCP-ZNN for joint-limited redundant manipulators to simultaneously achieve obstacle avoidance (OA) and joint limit avoidance (JLA);
A Pendulum-Driven Legless Rolling Jumping Robot
Jake Buzhardt, Phanindra Tallapragada
Robotic Intelligence
🎯 What it does: Designed and manufactured a legless rolling jumping robot that utilizes a drive arm to adjust weight distribution, and experimentally verified its high-speed rolling and jumping performance.
A Perching and Tilting Aerial Robot for Precise and Versatile Power Tool Work on Vertical Walls
Roman Dautzenberg, Rik Bähnemann
Robotic Intelligence
🎯 What it does: Developed an aerial robot integrating a quadrotor, active suction cups, and a lightweight tiltable linear tool platform, capable of performing precise and flexible electric tool operations on vertical walls.
A Physically Based Deformable Model with Haptic Feedback for Real-Time Robotic Surgery Simulation
S. Heredia, Yohei Kuroda
Robotic IntelligenceBiomedical DataPhysics Related
🎯 What it does: Proposed a physics-based deformable model for real-time laparoscopic cholecystectomy surgery simulation with the addition of tactile feedback
A Pretouch Perception Algorithm for Object Material and Structure Mapping to Assist Grasp and Manipulation Using a DMDSM Sensor
Fengzhi Guo, Dezhen Song
Robotic IntelligenceUltrasound
🎯 What it does: Propose a novel material and structure mapping (MSM) algorithm to assist robots in grasping and manipulation;
A Relative Infrastructure-less Localization Algorithm for Decentralized and Autonomous Swarm Formation
Dominik Schindler, Michele Magno
Robotic IntelligenceSimultaneous Localization and Mapping
🎯 What it does: Proposed an infrastructure-free relative localization algorithm suitable for decentralized autonomous drone swarm formation;
A Retractable Soft Growing Robot with a Flexible Backbone
Xinyi Pi, Lin Cao
Robotic Intelligence
🎯 What it does: Proposed and implemented a foldable soft growth robot, utilizing a scalable tubular backbone to achieve buckle-free retraction, and experimentally validated its working principle.
A Robotic Assistance Personalization Control Approach of Hip Exoskeletons for Gait Symmetry Improvement
Qiang Zhang, H. Huang
OptimizationRobotic IntelligenceReinforcement Learning
🎯 What it does: Designed and implemented a hierarchical control framework for bilateral hip exoskeletons to provide adaptive optimal assistance and achieve improved gait symmetry.
A Robust and Constrained Multi-Agent Reinforcement Learning Electric Vehicle Rebalancing Method in AMoD Systems
Sihong He, Fei Miao
OptimizationReinforcement Learning
🎯 What it does: Designed a robust and constrained multi-agent reinforcement learning framework and algorithm for electric vehicle scheduling balance in AMoD systems, considering the uncertainty of state transition kernels.
A ROS-Based Kinematic Calibration Tool for Serial Robots
Caroline Pascal, Alexandre Chapoutot
Robotic Intelligence
🎯 What it does: Developed a ROS-based modular tool for the calibration of forward kinematics in serial robots.
A Rotor Flywheel Robot: Land-air Amphibious Design and Control
Chunzheng Wang, Yangmin Li
Robotic Intelligence
🎯 What it does: Proposed a rotor flywheel robot with a single-wheel structure and eight rotors, achieving more flexible ground and aerial movement; built a dynamic model using the Kane method; designed a multi-input decoupling controller, and verified the robot's mode switching and rolling stability during ground movement through simulation and experiments.
A Safety Filter for Realizing Safe Robot Navigation in Crowds
Kaijun Feng, Yunjiang Lou
Robotic Intelligence
🎯 What it does: Developed a safety filter (SF) to enable safe navigation for mobile robots in crowded environments and provide safety guarantees for navigation behaviors;
A Sensitivity Analysis of an Economic Value Metric for Quantifying the Success of Lower-Limb Exoskeletons and Their Assistance
Nundini D. Rawal, Elliott J. Rouse
Robotic Intelligence
🎯 What it does: Explored the use of an economic value metric (Marginal Value) based on the Vickrey second-price auction to assess the assistive value of lower-limb exoskeletons, and conducted sensitivity analysis through simulating robot bidders.
A Shared Autonomous Nursing Robot Assistant with Dynamic Workspace for Versatile Mobile Manipulation
Nikita Boguslavskii, Zhi Li
Robotic Intelligence
🎯 What it does: A shared autonomous mobile humanoid care robot is proposed, equipped with a drivable multifunctional support structure that can automatically adjust the workspace to enhance reachability and manipulability in cluttered environments, and remote care tasks are simulated and evaluated through an open-source ROS-Unity virtual test platform.
A Shared-Control Dexterous Robotic System for Assisting Transoral Mandibular Fracture Reduction: Development and Cadaver Study
Yan Wang, K. W. S. Au
Robotic Intelligence
🎯 What it does: Designed and validated a shared control dexterous robotic system to assist in the decompression and fixation of mandibular fractures via the oral route.
A Single-Parameter Model for Soft Bellows Actuators under Axial Deformation and Loading
Emma Treadway, R. B. Gillespie
Robotic IntelligencePhysics Related
🎯 What it does: Proposed and validated a simple two-port model to describe the relationship between fluid and mechanical variables of a soft corrugated spring actuator under axial deformation and load, capturing elastic energy and evaluated experimentally.
A Small Form Factor Aerial Research Vehicle for Pick-and-Place Tasks with Onboard Real-Time Object Detection and Visual Odometry
Cora A. Dimmig, Marin Kobilarov
ClassificationObject DetectionObject TrackingPose EstimationSimultaneous Localization and Mapping
🎯 What it does: Propose a small aerial platform along with its hardware and software components for object detection, classification, tracking, and interaction, supporting safe and reliable grasping.
A Smart Handheld Edge Device for on-Site Diagnosis and Classification of Texture and Stiffness of Excised Colorectal Cancer Polyps
Ozdemir Can Kara, F. Alambeigi
ClassificationConvolutional Neural NetworkBiomedical Data
🎯 What it does: Proposed an intelligent handheld edge device equipped with a tactile sensing module and a two-stage machine learning algorithm for on-site diagnosis and classification of texture and stiffness of resected colorectal polyps.
A Soft, Multi-Layer, Kirigami Inspired Robotic Gripper with a Compact, Compression-Based Actuation System
Joao Buzzatto, Minas V. Liarokapis
Robotic Intelligence
🎯 What it does: Designed and tested a kirigami-based robotic gripper using compression-driven actuation, investigated the impact of internal cut shapes on grip strength and grasping capability, and validated its effectiveness in daily tasks through experiments
A Stable Adaptive Extended Kalman Filter for Estimating Robot Manipulators Link Velocity and Acceleration
Seyed Ali Baradaran Birjandi, Sami Haddadin
Robotic IntelligenceTime Series
🎯 What it does: A stable adaptive extended Kalman filter that can adaptively handle noise in real-time is proposed for estimating the velocity and acceleration of robot manipulator links through model-based sensor fusion (IMU + motor encoders).
A Tailsitter UAV Based on Bioinspired, Tendon-Driven, Shape-Morphing Wings with Aerofoil-Shaped Artificial Feathers
Junbang Liang, Minas V. Liarokapis
Robotic IntelligencePhysics Related
🎯 What it does: Designed and built a bio-inspired tail-sitting drone with morphing wing surfaces and airfoil-shaped artificial feathers, achieving full tendon-driven actuation and wing expansion/contraction controlled by a single motor per side; validated aerodynamic performance and low power consumption through wind tunnel tests and outdoor flights.
A Teleoperated MR-Safe Haptic System for Magnetic Resonance Imaging-Guided Prostate Needle Biopsies
Evelyn Mendoza, J. P. Whitney
Robotic IntelligenceBiomedical DataMagnetic Resonance Imaging
🎯 What it does: Designed and tested a passive MR-safe haptic teleoperation device for MRI-guided prostate needle biopsy.
A Telescopic Tendon-Driven Needle Robot for Minimally Invasive Neurosurgery
Saeed Rezaeian, Jun Sheng
Robotic IntelligenceBiomedical Data
🎯 What it does: Designed, modeled, and tested a flexible, telescopic tension-driven needle-shaped robot for minimally invasive neurosurgery.
A Thousand Worlds: Scenery Specification and Generation for Simulation-Based Testing of Mobile Robot Navigation Stacks
Samuel Parra, N. Hochgeschwender
Robotic IntelligenceWorld Model
🎯 What it does: Developed a domain-specific language and associated tools tailored for indoor environments to describe scenes and automatically generate variations, enabling simulation-based testing of mobile robot navigation stacks and reproduction of erroneous behaviors; an 8-year-old hidden bug in the ROS navigation stack was discovered during testing.
A Two-Dimensional Reticular Core Optical Waveguide Sensor for Tactile and Positioning Sensing
Zeyu Liu, Long Cheng
OptimizationPhysics Related
🎯 What it does: Designed and experimentally verified a two-dimensional lattice-core optical waveguide tactile and positioning sensor, studied its optical propagation characteristics and geometric parameters, and utilized a deep neural network to optimize spatial resolution.
A Two-Stage Based Social Preference Recognition in Multi-Agent Autonomous Driving System
Jintao Xue, Eryun Liu
RecognitionAutonomous DrivingReinforcement Learning
🎯 What it does: Proposes a two-stage system framework to enhance collaborative driving performance in multi-agent autonomous driving systems by sharing and identifying social value orientation (SVO).
A Unified Controller for Natural Ambulation on Stairs and Level Ground with a Powered Robotic Knee Prosthesis
Marissa Cowan, Tommaso Lenzi
Robotic Intelligence
🎯 What it does: Propose a unified controller that enables an electric knee prosthesis to seamlessly switch between walking, ascending stairs, and descending stairs without requiring explicit activity classification;
A Unified Perspective on Multiple Shooting In Differential Dynamic Programming
He Li, Patrick M. Wensing
OptimizationRobotic Intelligence
🎯 What it does: Extend DDP with multiple shooting, propose new derivations, penalty methods, adaptive merit functions, and more reliable acceptance conditions, and validate them in trajectory optimization for quadrotors, dual-arm robots, and manipulators, as well as in MPC for dynamic jumping of quadruped robots.
A Unified Trajectory Generation Algorithm for Dynamic Dexterous Manipulation
Cheng Zhou, Yu Zheng
OptimizationRobotic Intelligence
🎯 What it does: Proposed an efficient multi-stage trajectory generation algorithm for dynamic dexterous manipulation tasks, which decomposes tasks such as throwing, capturing, dynamic regrasping, and dynamic handover into manipulation primitives including adhesion, rolling, approaching, separation, collision, and grasping. Each primitive is modeled as a free-end optimal control problem, solving for optimal pose trajectories of the object and robot under pose and force coupling constraints; the algorithm's effectiveness was validated through single-arm regrasping and dual-arm dynamic handover experiments.
A Visibility-Based Escort Problem
Lance G. Fletcher, Jason M. O'Kane
Optimization
🎯 What it does: Proposed and solved a visibility-based escort planning problem, providing a complete path planning algorithm for the escort agent and VIP in a 2D environment, ensuring that the VIP is never directly visible to enemies at any time and safely reaches the target;
A Wearable Robotic Rehabilitation System for Neuro-Rehabilitation Aimed at Enhancing Mediolateral Balance
Zhenyuan Yu, H. Huang
Robotic IntelligenceBiomedical Data
🎯 What it does: Developed a neurorehabilitation platform integrating a wearable robotic hip abduction-adduction exoskeleton with a visual interface, and conducted movement training on three non-disabled subjects, analyzing its adaptability to hip joint kinematics and dynamics.
Accelerating Reinforcement Learning for Autonomous Driving Using Task-Agnostic and Ego-Centric Motion Skills
Tong Zhou, Y. Liu
Autonomous DrivingRepresentation LearningReinforcement Learning
🎯 What it does: Proposed a RL algorithm called TaEc-RL based on task-agnostic, ego-centric motion skill libraries, accelerating adaptive driving learning by exploring all feasible motion skills in continuous space.
Accessible Soft Robotics Education with Re-Configurable Balloon Robots
Yi-Shiun Wu, Josie Hughes
Robotic Intelligence
🎯 What it does: Developed a low-cost, reconfigurable balloon robot kit that can be quickly assembled and demonstrate key principles of soft robotics, such as morphology, stiffness control, controller dependency, and interaction with the environment.
Accurate and Interactive Visual-Inertial Sensor Calibration with Next-Best-View and Next-Best-Trajectory Suggestion
Christopher Choi, Stefan Leutenegger
Simultaneous Localization and MappingMultimodality
🎯 What it does: Proposes a calibration process for visual-inertial (VI) sensors, utilizing a graphical user interface (GUI) and information theory to guide non-experts in collecting information-rich calibration data through recommendations for the next best view and next best trajectory, enabling calibration of intrinsic parameters, extrinsic parameters, and time errors.
AcouSkin: Full Surface Contact localization Using Acoustic Waves
A. Kosta, Volkan Isler
Robotic IntelligenceAudio
🎯 What it does: Proposed a full-surface touch localization system called AcouSkin based on acoustic waves
Acquisition and Prediction of High-Density Tactile Field Data for Rigid and Flexible Objects
Hongxiang Xue, Fuchun Sun
Data SynthesisRobotic IntelligenceImage
🎯 What it does: Proposes a high-density tactile field data acquisition method based on a mechanical simulation environment and constructs a network to learn the mapping between tactile images and six-dimensional tactile field information;
AcTag: Opti-Acoustic Fiducial Markers for Underwater Localization and Mapping
Kalin Norman, Joshua G. Mangelson
Object DetectionSimultaneous Localization and MappingImage
🎯 What it does: Designed and implemented AcTag acoustic tags and their detection method, evaluating their performance in both simulated and real sonar images.
Active Acoustic Sensing for Robot Manipulation
Shihan Lu, Heather Culbertson
Robotic IntelligenceAudio
🎯 What it does: Developed an active acoustic sensing method that utilizes vibration actuators and piezoelectric microphones to capture the resonance characteristics of objects for robot manipulation tasks.
Active Capsule System for Multiple Therapeutic Patch Delivery: Preclinical Evaluation
Jihun Lee, Sukho Park
Robotic IntelligenceDrug Discovery
🎯 What it does: Designed and verified an active capsule system capable of moving within the gastrointestinal tract, real-time imaging, and actively delivering multiple therapeutic patches, completing functional validation, ex vivo multi-patch delivery, and preclinical trials in pigs.
Active Electric Perception-Based Haptic Modality with Applications to Robotics
Michael Zechmair, Yannick Morel
Robotic Intelligence
🎯 What it does: This paper implements and characterizes a capacitive sensor for detecting and locating nearby objects, which can be mounted on any robot exterior. It is particularly suitable for detecting capacitive materials such as living tissue, and its application in promoting human-robot interaction is verified in interactive scenarios such as following and avoidance.
Active Planar Mass Distribution Estimation with Robotic Manipulation
Jiacheng Yuan, Volkan Isler
Robotic Intelligence
🎯 What it does: Estimating the planar mass distribution of a rigid body through robot interaction and force/torque feedback
Active Task Randomization: Learning Robust Skills via Unsupervised Generation of Diverse and Feasible Tasks
Kuan Fang, Jeannette Bohg
GenerationRobotic IntelligenceGraph Neural NetworkImage
🎯 What it does: Propose Active Task Randomization (ATR), which generates diverse and feasible training tasks in an unsupervised manner, learning robust manipulation skills applicable across multiple environments. These skills can be combined by a task planner to solve unknown sequential manipulation problems based on visual inputs.
Active Visual SLAM Based on Hierarchical Reinforcement Learning
Wensong Chen, Yu Hu
Robotic IntelligenceReinforcement LearningSimultaneous Localization and Mapping
🎯 What it does: Proposed a modular active visual SLAM system AVS-HRL based on hierarchical reinforcement learning, designed a dedicated map reconstruction module to correct cumulative errors in incremental mapping, and used abstract information (e.g., grid maps) as neural network inputs; conducted experiments in two simulators and real environments; in the Habitat noisy environment, improved mapping accuracy by 68.48% on average for the Gibson and MP3D datasets; demonstrated direct transfer capability across simulators and real environments.
Adaptive and Fail-Safe Magnetic Gripper with Charging Function for Drones on Power Lines
Viet Duong Hoang, E. Ebeid
Safty and PrivacyRobotic IntelligencePhysics Related
🎯 What it does: Developed an adaptive, lightweight, and safely releasable magnetic gripper suitable for drones to grasp and charge on power lines.
Adaptive Exploration-Exploitation Active Learning of Gaussian Processes
George P. Kontoudis, Michael W. Otte
Optimization
🎯 What it does: Proposed an adaptive exploration-exploitation active learning method ALX suitable for Gaussian Process surrogates, using an acquisition function that maximizes approximate expected Fisher information during the exploration phase, and a closed-form acquisition function that maximizes total expected variance reduction during the exploitation phase; the switch between exploration and exploitation phases is determined by prediction accuracy metrics; this method can execute quickly and is suitable for real-time decision making.
Adaptive PD Control Using Deep Reinforcement Learning for Local-Remote Teleoperation with Stochastic Time Delays
Lucy McCutcheon, Saber Fallah
OptimizationRobotic IntelligenceReinforcement Learning
🎯 What it does: Propose an adaptive PD control method using deep reinforcement learning to address random time delay issues in local-remote collaborative systems
Adaptive Robust Model Predictive Control for Bilateral Teleoperation
Fadi Alyousef Almasalmah, Bernard Bayle
OptimizationRobotic Intelligence
🎯 What it does: An adaptive robust model predictive controller (MPC) for a bidirectional teleoperation system was developed, incorporating a polytopic tube controller in the MPC design to ensure robust constraint satisfaction, and using an online learning method to learn the environment model;
AdaptSeqVPR: An Adaptive Sequence-Based Visual Place Recognition Pipeline
Heshan Li, Danwei Wang
RetrievalAutonomous DrivingConvolutional Neural NetworkSimultaneous Localization and MappingImageSequential
🎯 What it does: Proposes an adaptive sequence visual localization system called AdaptSeqVPR for vehicles traveling at non-uniform speeds, which includes an adaptive sequence search strategy named AdaptSeq and a CNN-based discriminator called DDsNet.
Addressing the Scale Shrinkage Problem in Learning-based Binocular Depth Estimation
Gang Guo, Fuchun Sun
Depth EstimationSupervised Fine-TuningImageBenchmark
🎯 What it does: Address the scale contraction problem in stereo depth estimation
Adjusting the Quasi-Stiffness of an Ankle-Foot Prosthesis Improves Walking Stability during Locomotion over Compliant Terrain
Chrysostomos Karakasis, Panagiotis K. Artemiadis
Robotic Intelligence
🎯 What it does: Proposed and evaluated a continuous impedance controller to adjust the quasi-stiffness of a controlled ankle-foot prosthesis, thereby enhancing gait stability on soft ground.
ADMNet: Anti-Drone Real-Time Detection and Monitoring
Xunkuai Zhou, Ben M. Chen
Object DetectionConvolutional Neural NetworkImage
🎯 What it does: Proposed a lightweight, effective, and efficient anti-drone network called ADMNet for real-time visual detection and monitoring of hostile drones in complex environments.
Adversarial Driving Behavior Generation Incorporating Human Risk Cognition for Autonomous Vehicle Evaluation
Zhen Liu, Xun Gong
Autonomous DrivingAdversarial AttackReinforcement Learning
🎯 What it does: This paper proposes a new framework for generating adversarial driving behaviors against background vehicles to expose effective and reasonable hazardous events for autonomous vehicles.
Adversarial Object Rearrangement in Constrained Environments with Heterogeneous Graph Neural Networks
Xibai Lou, Changhyun Choi
Robotic IntelligenceConvolutional Neural NetworkGraph Neural Network
🎯 What it does: Developed a hierarchical robot manipulation system for rearranging adversarial objects in constrained environments.
Aerial Manipulator Systems Gain a New Skill: Achieve Contact-based Landing on a Mobile Platform
Xiangdong Meng, Aiguo Song
Robotic Intelligence
🎯 What it does: Researched and implemented contact landing for drones and robotic arms, designed a robust controller, and verified its reliability and repeatability through experiments.