IROS 2023 Papers — Page 3
IEEE/RSJ International Conference on Intelligent Robots and Systems · 1195 papers
CoFlyers: A Universal Platform for Collective Flying of Swarm Drones
Jialei Huang, Tianjiang Hu
Robotic Intelligence
🎯 What it does: Proposed and developed an open-source platform named CoFlyers for end-to-end development from cluster models to real-world drone swarm flights.
Cognition Difference-Based Dynamic Trust Network for Distributed Bayesian Data Fusion
Yingke Li, Fumin Zhang
🎯 What it does: Proposed a mechanism based on perception differences for constructing a dynamic trust network for real-time distributed Bayesian data fusion (DDF)
Cognitive Approach to Hierarchical Task Selection for Human-Robot Interaction in Dynamic Environments
Syed T. Bukhari, W. Qazi
Robotic Intelligence
🎯 What it does: In dynamic environments, task selection for human-robot interaction is achieved through semantic memory and hierarchical task trees, enabling the robot to automatically perform appropriate tasks such as making tea or sandwiches based on implicit user cues and environmental context.
Collaborative Trolley Transportation System with Autonomous Nonholonomic Robots
Bingyi Xia, M. Meng
Robotic Intelligence
🎯 What it does: Developed an autonomous system for non-homogeneous multi-robot collaborative transportation of luggage carts, proposing an end-to-end hierarchical autonomous framework capable of real-time feasible path planning, motion plan generation, and handling long queues of luggage carts, obstacles, and pedestrians in complex dynamic environments through feedback control.
Collective Decision-Making and Change Detection with Bayesian Robots in Dynamic Environments
Kai Pfister, H. Hamann
ClassificationRobotic Intelligence
🎯 What it does: Proposed a control algorithm that combines statistical change detection with Bayesian robots for binary classification of distributed features in dynamic environments.
Collision Isolation and Identification Using Proprioceptive Sensing for Parallel Robots to Enable Human-Robot Collaboration
Aran Mohammad, T. Ortmaier
Safty and PrivacyRobotic Intelligence
🎯 What it does: A collision isolation and identification algorithm based on proprioception was developed and verified in parallel robots to enhance human-robot collaboration safety.
Collision Prevention Strategy Using Sparse 2D Spatial Information for Indoor Mobile Robots
Youngil Koh, MidEum Choi
Robotic IntelligencePoint Cloud
🎯 What it does: This paper proposes a collision avoidance strategy for indoor mobile robots based on low-cost 2D laser scanners.
Collision-Free Reconfiguration Planning for Variable Topology Trusses Using a Linking Invariant
Alexander Spinos, Mark Yim
Robotic Intelligence
🎯 What it does: Proposes a multimodal reconfiguration planner applicable to Variable Topology Truss (VTT) modular robotic systems
Combinatorial Disjunctive Constraints for Obstacle Avoidance in Path Planning
Raul Garcia, Joey Huchette
Optimization
🎯 What it does: Proposes a novel method for modeling obstacle avoidance constraints in 2D environments by assigning waypoints to obstacle-free polyhedral regions, utilizing combinatorial discriminant constraints and independent branching to construct a compact and ideal mixed-integer programming (MIP) model.
Combining Measurement Uncertainties with the Probabilistic Robustness for Safety Evaluation of Robot Systems
W.-J. Baek, Torsten Kröger
Safty and PrivacyRobotic Intelligence
🎯 What it does: Propose a method that combines measurement uncertainty with probabilistic robustness to quantify uncertainty in robotic systems online and evaluate safety thresholds.
Communicating human intent to a robotic companion by multi-type gesture sentences
Petr Vanc, Václav Hlaváč
Pose EstimationRobotic IntelligenceVision-Language-Action ModelImage
🎯 What it does: Propose a multi-type gesture pseudo-language, using gesture combinations to express human intentions, and implement real-time intent extraction and robot response based on behavior trees, supporting switching from gesture commands to direct end-effector operations, evaluated using the Franka Emika Panda 7-DoF robotic arm.
Communication Resources Constrained Hierarchical Federated Learning for End-to-End Autonomous Driving
Wei-Bin Kou, Yik-Chung Wu
Autonomous DrivingOptimizationFederated Learning
🎯 What it does: Proposed an optimized communication resource constrained hierarchical federated learning framework (CRCHFL) that minimizes generalization error in autonomous driving models through hybrid data and model aggregation.
Compact and Accurate Adaptive Width Radial Basis Function Neural Network with Discriminative Features for Thyroid Gland Segmentation
Haijun You, Junying Chen
SegmentationImageBiomedical DataUltrasound
🎯 What it does: Developed a fast, compact, accurate, and robust adaptive-width radial basis function neural network for automatically segmenting thyroid ultrasound images in resource-constrained environments.
Comparing Quadrotor Control Policies for Zero-Shot Reinforcement Learning under Uncertainty and Partial Observability
Sven Gronauer (Technische Universität Darmstadt), K. Diepold (Technische Universität Darmstadt)
Robotic IntelligenceRecurrent Neural NetworkReinforcement LearningSequential
🎯 What it does: Investigate historical dependency strategies for drone control in zero-shot training environments and validate them on real quadrotors.
Comparing the Effectiveness of Control Methodologies of a Hip-Knee-Ankle Exoskeleton During Squatting
Jared M. Li, Gwendolyn M. Bryan
🎯 What it does: Studied the effectiveness of two control methods (virtual model control and gravity compensation) in reducing lower limb muscle activity in hip-knee-ankle exoskeletons.
Competitive Ant Coverage: The Value of Pursuit
Alon Shats, Noa Agmon
Robotic Intelligence
🎯 What it does: Studies the competitive strategies of two ants-like robots with limited capabilities in a regional coverage task, comparing FCC and LCC objectives.
COMPlacent: A Compliant Whisker Manipulator for Object Tactile Exploration
Chenxi Xiao, J. Wachs
Robotic IntelligenceTime Series
🎯 What it does: Proposed a COMPlacent flexible whisker-like manipulator for achieving minimally invasive tactile exploration when contacting fragile objects.
Complete Coverage Path Planning for Omnidirectional Expand and Collapse Robot Panthera
Lim Yi, M. R. Elara
OptimizationRobotic Intelligence
🎯 What it does: Effective area coverage methods using an omni-directional expandable/contractable robot (OECR) were studied, and a depth-first search (DFS) approach was proposed to achieve complete coverage path planning.
CompUDA: Compositional Unsupervised Domain Adaptation for Semantic Segmentation Under Adverse Conditions
Ziqiang Zhengl, Sai-Kit Yeung
SegmentationDomain AdaptationImage
🎯 What it does: This paper proposes an unsupervised domain adaptation method called CompUDA based on multi-factor decomposition for semantic segmentation under adverse weather conditions. The method improves semantic segmentation performance by decomposing domain differences into three factors: style, visibility, and image quality, and adapting them individually using intermediate domains.
Computational Design of Closed-Chain Linkages: Respawn Algorithm for Generative Design
D.V. Ivolga, S. Kolyubin
GenerationOptimization
🎯 What it does: Proposed an optimization-driven algorithm for generating planar closed-chain linkage mechanisms that satisfy predefined trajectories
Computing Motion Plans for Assembling Particles with Global Control
Patrick Blumenberg, Aaron T. Becker
Optimization
🎯 What it does: Studies the motion planning algorithm for unit square tiles in a grid world under uniform external forces within the tilt model, used for shape assembly.
Consecutive Inertia Drift of Autonomous RC Car via Primitive-Based Planning and Data-Driven Control
Yiwen Lu, Yilin Mo
Autonomous Driving
🎯 What it does: Proposed a continuous inertia drift scheme for RC cars based on primitive planning and data-driven control
ConSOR: A Context-Aware Semantic Object Rearrangement Framework for Partially Arranged Scenes
Kartik Ramachandruni, Sonia Chernova
🎯 What it does: Proposes ConSOR, which rearranges objects by leveraging contextual cues from partially arranged scenes without requiring users to explicitly specify the target scene;
Constrained Dynamic Movement Primitives for Collision Avoidance in Novel Environments
S. Shaw, D. Nikovski
OptimizationRobotic Intelligence
🎯 What it does: Proposed a Constrained Dynamic Movement Primitives (CDMP) that can satisfy workspace position constraints
Constrained Generative Sampling of 6-DoF Grasps
Jens Lundell, Ville Kyrki
GenerationPose EstimationPoint Cloud
🎯 What it does: Proposed a generative network VCGS for constrained 6-DoF grasp sampling, and constructed a new constrained grasp dataset CONG.
Constrained Reinforcement Learning and Formal Verification for Safe Colonoscopy Navigation
Davide Corsi, D. Dall’Alba
Reinforcement LearningBiomedical Data
🎯 What it does: Proposed a colonoscopy navigation strategy based on deep reinforcement learning (DRL), eliminating the need for lumen tracking and ensuring the safety of the navigation strategy through constrained reinforcement learning (CRL) and formal verification (FV); validated in a virtual colonoscopy environment, ultimately selecting three completely safe strategies from 300 trained strategies.
Constraint Programming for Component-Level Robot Design
Andrew Wilhelm, Nils Napp
OptimizationRobotic Intelligence
🎯 What it does: Model the robot component selection problem as a combinatorial optimization problem, and solve it using the depth-first branch and bound algorithm of constraint programming, enhanced with two specific heuristic strategies and redundant constraints to improve solving efficiency.
Constructing Metric-Semantic Maps Using Floor Plan Priors for Long-Term Indoor Localization
Nicky Zimmerman, C. Stachniss
Object DetectionRobotic IntelligenceSimultaneous Localization and MappingImage
🎯 What it does: Utilize 3D object detection from monocular RGB images to construct a metric-semantic map and achieve long-term goal-based localization on this map.
Contact Reduction with Bounded Stiffness for Robust Sim-to-Real Transfer of Robot Assembly
Nghia Vuong, Quang Pham
Domain AdaptationRobotic IntelligenceReinforcement LearningMesh
🎯 What it does: Proposed and implemented a contact point simplification method with bounded stiffness to improve simulation accuracy and support transfer from simulation to real-world reinforcement learning.
Contact-Aware Non-Prehensile Manipulation for Object Retrieval in Cluttered Environments
Yongpeng Jiang, Xiang Li
OptimizationRobotic Intelligence
🎯 What it does: Propose a non-grasping object retrieval method based on a rod-like actuator, which can generate candidate trajectories first and then complete retrieval by using MPC control to adjust the slider's posture in cluttered environments
Contact-Aware Shaping and Maintenance of Deformable Linear Objects With Fixtures
Ke Chen, A. Knoll
Robotic IntelligenceImage
🎯 What it does: Proposes a framework that utilizes environmental contact to control and maintain the shape of deformable linear objects using two robotic arms.
Contactless Weight Estimation of Human Body and Body Parts for Safe Robotics-Assisted Casualty Extraction
Jaeyeon Lee, Nathan Fisher
Robotic IntelligenceConvolutional Neural NetworkImagePoint Cloud
🎯 What it does: Developed a vision-based non-contact method to estimate the weight of the human body and its individual parts, enabling robots to safely extract casualties from hazardous environments.
Content Estimation Through Tactile Interactions with Deformable Containers
Yueyang Liu, Shiao-Li Tsao
ClassificationData-Centric LearningRobotic IntelligenceTime SeriesSequential
🎯 What it does: Proposed a learning-based model that estimates content properties of different materials and heights by utilizing tactile data collected when gently pressing containers.
Continuous Implicit SDF Based Any-Shape Robot Trajectory Optimization
Ting Zhang, Fei Gao
OptimizationRobotic Intelligence
🎯 What it does: Propose a trajectory optimization method for arbitrary-shaped robots based on continuous implicit SDF
Control of Cart-Like Nonholonomic Systems Using a Mobile Manipulator
Sergio Aguilera, Seth Hutchinson
OptimizationRobotic Intelligence
🎯 What it does: Dynamic modeling of a nonholonomic system similar to a shopping cart using a mobile manipulator, LQR trajectory tracking, and control law design, with experimental validation of improved trajectory tracking accuracy and controllability.
Control Transformer: Robot Navigation in Unknown Environments Through PRM-Guided Return-Conditioned Sequence Modeling
Daniel Lawson, A. H. Qureshi
Robotic IntelligenceTransformer
🎯 What it does: Proposed and implemented Control Transformer for long-distance navigation in unknown environments, combined with PRM-guided return conditional sequence modeling
Controller and Trajectory Optimization for a Quadrotor UAV with Parametric Uncertainty
A. Srour, P. Giordano
Autonomous DrivingOptimization
🎯 What it does: This paper utilizes the concept of closed-loop state sensitivity to compare three typical quadrotor controllers and proposes an optimization problem that simultaneously considers the reference trajectory shape and controller gain. Subsequently, large-scale statistical experiments are conducted to evaluate its impact on closed-loop robustness under model parameter uncertainty.
Controlling Powered Prosthesis Kinematics over Continuous Transitions Between Walk and Stair Ascent
Shihao Cheng, Robert D. Gregg
Robotic IntelligenceBiomedical Data
🎯 What it does: Propose a phase-based joint motion controller to enable smooth, biomimetic transitions between walking and stair climbing for electric knee-ankle prosthetics.
Converting Depth Images and Point Clouds for Feature-Based Pose Estimation
Robert Lösch, Bernhard Jung
Image TranslationPose EstimationSimultaneous Localization and MappingImagePoint Cloud
🎯 What it does: Proposed a new method to convert depth images and point clouds into images that can visualize spatial details, which encodes information by calculating the difference between two normal vectors of neighboring points after denoising.
Convolutional Occupancy Models for Dense Packing of Complex, Novel Objects
Nikhil Mishra, Maximilian Sieb
OptimizationRepresentation LearningData-Centric LearningRobotic IntelligenceConvolutional Neural Network
🎯 What it does: Propose a fully convolutional shape completion model called F-CON, combine it with existing planning methods to achieve dense packing of complex, unseen objects in real pick-and-place systems; simultaneously release the simulated dataset COB-3D-v2;
Cooperative Dual-Arm Control for Heavy Object Manipulation Based on Hierarchical Quadratic Programming
Maximilian Dio, Knut Graichen
OptimizationRobotic Intelligence
🎯 What it does: Proposes a dual-arm collaborative heavy object manipulation control scheme based on hierarchical quadratic programming (HQP), utilizing a full dynamic model to achieve dynamic inequality constraints, optimizing trajectory tracking and load distribution.
Cooperative LiDAR Localization and Mapping for V2X Connected Autonomous Vehicles
Bingyi Cao, Daniel Goehring
Autonomous DrivingSimultaneous Localization and MappingPoint Cloud
🎯 What it does: Proposed and implemented a CAV cooperative SLAM system using sparse LiDAR feature representation, which enables data exchange between vehicles through V2X messages.
Coordination of Bounded Rational Drones Through Informed Prior Policy
Durgakant Pushp, Lantao Liu
OptimizationRobotic IntelligenceReinforcement Learning
🎯 What it does: Designed a prior strategy that leverages goal-oriented navigation heuristics in familiar environments and adapts to unfamiliar environments through environment-dependent exploration noise, achieving collaborative motion planning for multi-robot systems under a bounded rationality framework; combined this prior strategy with the bounded rationality framework from game theory to enhance group decision-making efficiency; and conducted experiments in simulations and real-world scenarios;
Coordination of Multiple Mobile Manipulators for Ordered Sorting of Cluttered Objects
Jeeho Ahn, Changjoo Nam
OptimizationRobotic Intelligence
🎯 What it does: A fast multi-robot collaboration scheme was developed, using a search algorithm to determine a sorting sequence that satisfies intra-group ordering constraints, and assigning objects to mobile manipulators through greedy task allocation.
Counterexample Guided Abstraction Refinement with Non-Refined Abstractions for Multi-Goal Multi-Robot Path Planning
Pavel Surynek
OptimizationRobotic Intelligence
🎯 What it does: Using the CEgar framework to solve the multi-goal multi-robot path planning (MG-MRPP) problem.
Covering Dynamic Demand with Multi-Resource Heterogeneous Teams
Mela C. Coffey, Alyssa Pierson
OptimizationRobotic Intelligence
🎯 What it does: The study focuses on teams of heterogeneous robots equipped with different types and quantities of resources, tasked with providing resources to multiple dynamic demand locations. It proposes an adaptive control strategy that depletes demand as robots supply resources, while allowing for demand injection and movement of demand locations.
Creating a Dynamic Quadrupedal Robotic Goalkeeper with Reinforcement Learning
Xiaoyu Huang, K. Sreenath
Robotic IntelligenceReinforcement Learning
🎯 What it does: Propose a hierarchical model-free reinforcement learning framework enabling quadruped robots to perform goalkeeper tasks, intercepting flying soccer balls within less than one second using dynamic locomotion actions.
CREPES: Cooperative RElative Pose Estimation System
Zhiren Xun, Yanjun Cao
Pose EstimationSimultaneous Localization and Mapping
🎯 What it does: Proposes the CREPES system to achieve 6-DOF relative pose estimation among multiple robots.
Cross-Domain Autonomous Driving Perception Using Contrastive Appearance Adaptation
Ziqiang Zheng, Sai-Kit Yeung
Domain AdaptationAutonomous DrivingContrastive LearningImage
🎯 What it does: Proposed an image-level domain adaptation framework that performs domain transfer through content-aligned source-target image pairs, and constructs a cyclic structure with three modules: cross-domain content alignment, reference-guided image synthesis, and contrastive learning.
CurveQuad: A Centimeter-Scale Origami Quadruped that Leverages Curved Creases to Self-Fold and Crawl with One Motor
Daniel Feshbach, Cynthia R. Sung
Robotic Intelligence
🎯 What it does: Designed, manufactured, and experimentally verified a miniature curved origami quadruped robot capable of performing multimodal actions such as self-folding, unfolding, crawling, and turning using a single actuator.
Cyber-Attacks on Wheeled Mobile Robotic Systems with Visual Servoing Control
Aleksandar Jokić, Miroslav Pajic
Adversarial AttackRobotic IntelligenceImageVideo
🎯 What it does: Conduct a safety analysis of the visual servo system in wheeled mobile robots and propose two types of deception attacks.
D-PALI: A Low-Cost Open Source Robotic Gripper Platform for Planar In-Hand-Manipulation
Arunansu Patra, Adam Spiers
Robotic Intelligence
🎯 What it does: Designed and implemented a low-cost, open-source dual-finger 2DOF five-bar linkage robotic hand for planar in-hand manipulation, supporting modular expansion to 3+ fingers.
Daily Assistive Modular Robot Design Based on Multi-Objective Black-Box Optimization
Kento Kawaharazuka, M. Inaba
OptimizationHyperparameter SearchRobotic Intelligence
🎯 What it does: A reconfigurable drive module was developed, featuring a locking mechanism for carrying heavy objects and a release mechanism for convenient manual teaching, and automatically generating modular robot configurations that satisfy multi-objective coordinates while minimizing torque using the Tree-structured Parzen Estimator.
DAMON: Dynamic Amorphous Obstacle Navigation using Topological Manifold Learning and Variational Autoencoding
Apan Dastider, Mingjie Lin
Robotic IntelligenceAuto Encoder
🎯 What it does: Achieve dynamic obstacle navigation and collision avoidance by learning a low-dimensional hierarchical manifold graph combined with a variational autoencoder, supporting motion planning on 7-DoF robotic arms.
DANDELION: An ASV Deployed Micro-Profiler Array for Air-Sea Observation
Zhihao Fan, Zheng Zeng
🎯 What it does: Developed and verified the DANDELION observation system, which consists of an autonomous floating vehicle (ASV), a launcher with high-speed friction wheels, and eight dandelion-style micro-sensors, with field tests conducted in Qiandaohu Lake.
Data-Based MHE for Agile Quadrotor Flight
Wonoo Choo, E. Kayacan
OptimizationRobotic Intelligence
🎯 What it does: Developed a data-based moving window estimation (MHE) method for state estimation in agile quadrotor flight.
Data-Driven Adaptation for Robust Bipedal Locomotion with Step-to-Step Dynamics
Min Dai, A. Ames
Data-Centric LearningRobotic Intelligence
🎯 What it does: Propose an online framework that synthesizes agile gaits for bipedal robots through data-driven S2S dynamics and adaptive control, enabling adaptation to unknown environments, model errors, and external disturbances.
Data-Driven Based Cascading Orientation and Translation Estimation for Inertial Navigation
Xiangyu Deng, Yandong Guo
Pose EstimationConvolutional Neural NetworkTransformerSimultaneous Localization and MappingTime Series
🎯 What it does: Propose a cascading pose and displacement estimation method based on IMU-only measurements.
Data-Driven Distributionally Robust Optimal Control with State-Dependent Noise
Rui Liu, Pratap Tokekar
OptimizationRobotic Intelligence
🎯 What it does: Proposes a data-driven distributionally robust optimal control method (D3ROC) that achieves robust control under unknown noise distributions.
Data-Driven Modeling and Experimental Validation of Autonomous Vehicles Using Koopman Operator: Distribution A: Approved for Public Release; Distribution Unlimited. OPSEC # 7248
Ajinkya Joglekar, V. Krovi
Autonomous Driving
🎯 What it does: Establish and verify a linear prediction model based on Koopman EDMD and MPC feedback control for trajectory tracking of a small F1TENTH vehicle.
Data-Efficient Online Learning of Ball Placement in Robot Table Tennis
Philip Tobuschat, Michael Muehlebach
OptimizationRobotic Intelligence
🎯 What it does: Implemented an online optimization algorithm for a robot table tennis machine to hit a predefined target when returning a ball.
Data-Efficient Policy Selection for Navigation in Partial Maps via Subgoal-Based Abstraction
Abhishek Paudel, Gregory J. Stein
Reinforcement Learning
🎯 What it does: Proposed a method for rapid and reliable partial map navigation strategy selection using offline all-strategy replay combined with learning-enhanced subgoal planning (LSP).
Decentralised Multi-Robot Exploration Using Monte Carlo Tree Search
Sean Bone, M. Chli
Robotic Intelligence
🎯 What it does: Propose a distributed multi-robot exploration method based on Monte Carlo Tree Search
Decentralized Connectivity Maintenance for Quadrotor UAVs with Field of View Constraints
M. Bernard, P. Giordano
OptimizationRobotic Intelligence
🎯 What it does: Proposed a decentralized connectivity maintenance algorithm addressing the field-of-view limitations and lack of a shared reference frame in quadrotor UAVs
Decentralized Multi-Agent Reinforcement Learning with Global State Prediction
Josh Bloom, Carlo Pinciroli
Robotic IntelligenceReinforcement Learning
🎯 What it does: The study addresses multi-robot collaboration scenarios lacking global information, achieving collective transportation tasks in multi-agent deep reinforcement learning through implicit communication or a Global State Prediction (GSP) network.
Decentralized Planning for Car-Like Robotic Swarm in Cluttered Environments
Changjia Ma, Fei Gao
OptimizationRobotic Intelligence
🎯 What it does: Proposed a distributed planning framework for vehicle-type robot swarms that can achieve real-time planning in complex environments.
Decentralized Swarm Trajectory Generation for LiDAR-based Aerial Tracking in Cluttered Environments
Longji Yin, Fu Zhang
OptimizationRobotic IntelligencePoint Cloud
🎯 What it does: Proposes a decentralized formation trajectory planner for LiDAR-based perception multi-robot aerial tracking, maximizing target visibility and avoiding collisions in crowded environments.
Decision-Oriented Learning with Differentiable Submodular Maximization for Vehicle Routing Problem
Guan-Yu Shi, Pratap Tokekar
Optimization
🎯 What it does: A learning method is studied to map environmental observations to parameters of a submodular function for the mobile charging station routing problem involving drones and unmanned ground vehicles.
Decoding sEMG Under Non-Ideal Conditions Toward Robust Muscle-Machine Interface Control
C. D. Guerrero-Méndez, R. M. Andrade
ClassificationRecognitionBiomedical Data
🎯 What it does: Evaluate the sEMG gesture recognition method based on Extreme Learning Machine under five non-ideal conditions
Deconfounded Opponent Intention Inference for Football Multi-Player Policy Learning
Shijie Wang, Jianqiang Yi
Graph Neural NetworkReinforcement LearningBenchmark
🎯 What it does: Proposes a Debiased Opponent Intent Inference (DOII) method for multi-player strategy learning in football; infers opponent intent through an Opponent Intent Supervision Module, eliminates the impact of interfering factors on causal relationships using a Debiased Trajectory Graph Module, and enhances players' sensitivity to opponent intent via an Opponent Reward Module; experiments show that the method effectively improves player strategy performance in the Google Research Football environment.
Deep Functional Predictive Control (deep-FPC): Robot Pushing 3-D Cluster Using Tactile Prediction
Kiyanoush Nazari, Amir Ghalamzan-E
Robotic IntelligenceWorld ModelSequentialAgriculture Related
🎯 What it does: Proposed a data-driven forward model based on tactile prediction, integrated into deep functional predictive control (d-FPC), for controlling the displacement of a robot fingertip when pushing tomato plants.
Deep Learning-Based Leaf Detection for Robotic Physical Sampling with P-AgBot
Aarya Deb, D. Cappelleri
SegmentationPose EstimationRobotic IntelligenceConvolutional Neural NetworkImageAgriculture Related
🎯 What it does: Propose a deep learning-based leaf detection and segmentation method, using an RGB-D camera to collect data and train Mask R-CNN and YOLOv5, while proposing a leaf pose estimation algorithm to maximize leaf sampling area, combined with the P-AgBot platform to realize robotic physical sampling.
Deep Probabilistic Movement Primitives with a Bayesian Aggregator
Michael Przystupa, Samuele Tosatto
Robotic Intelligence
🎯 What it does: Proposed a unified deep motion primitive architecture that supports time modulation, mixing, point constraints, and context conditioning, and improved context conditioning and mixing using a Bayesian aggregator.
Deep Reinforcement Learning-Based Intelligent Traffic Signal Controls with Optimized CO2 Emissions
Pedram Agand, Mo Chen
Autonomous DrivingOptimizationReinforcement Learning
🎯 What it does: Designed and proposed the EcoLight reward shaping scheme for optimizing traffic signal control using deep reinforcement learning algorithms, aiming to reduce CO2 emissions while balancing with travel time and other indicators.
Deep Robust Multi-Robot Re-Localisation in Natural Environments
Milad Ramezani, Peyman Moghadam
Robotic IntelligenceSimultaneous Localization and MappingImagePoint Cloud
🎯 What it does: Propose a self-supervised 2D-3D feature matching method based on LiDAR and image cross-modal data, using a deep network to extract LiDAR features and estimate relative poses between point clouds to achieve robust relocalization in natural environments.
Demonstrating Autonomous 3D Path Planning on a Novel Scalable UGV-UAV Morphing Robot
Eric Sihite, M. Gharib
Robotic IntelligenceSimultaneous Localization and Mapping
🎯 What it does: Designed and implemented a deformable multi-mode robot capable of moving on the ground and in the air, and achieved autonomous navigation in complex indoor environments using off-the-shelf SLAM and path planning algorithms.
Depth Self-Supervision for Single Image Novel View Synthesis
Giovanni Minelli, Samuele Salti
GenerationDepth EstimationImage
🎯 What it does: Proposes a method to generate new images from arbitrary perspectives under single-frame input, simultaneously optimizing novel view synthesis and depth estimation through a shared self-supervised depth decoder.
Depth-Based 6DoF Object Pose Estimation Using Swin Transformer
Zhujun Li, I. Stamos
Pose EstimationConvolutional Neural NetworkTransformerImagePoint Cloud
🎯 What it does: Proposes the SwinDePose framework, which utilizes the geometric information of depth images to achieve accurate 6D pose estimation.
Design and Control of a Ballbot Drivetrain with High Agility, Minimal Footprint, and High Payload
Chenzhang Xiao, E. Hsiao-Wecksler
OptimizationRobotic Intelligence
🎯 What it does: Designed and controlled a spherical robot transmission system with high agility, minimal footprint, and high load capacity, constructed full-scale prototype MiaPURE and simplified test platform PIPTB, and proposed a robust cascaded LQR-PI controller.
Design and Control of a Reluctance-Based Micropositioning Stage for Scanning Motion Applications
M. A. Saaideh, M. Janaideh
Physics Related
🎯 What it does: This paper designs and characterizes a magnetic reluctance-driven micro-positioning stage, including CAD modeling, prototype manufacturing, mathematical modeling, and experiments on force-current and force-gap characteristics.
Design and Control of Microscale Dual Locomotion Mode Multi-Functional Robots (μDMMFs)
Aaron C. Davis, D. Cappelleri
Robotic IntelligencePhysics Related
🎯 What it does: Designed and controlled a micro-scale dual-mode multifunctional robot that employs two magnetic actuation modes (rotational field and gradient field), enabling precise micro-object manipulation and equipped with multiple end-effectors.
Design and Development of a Deformable In-Pipe Inspection Robot for Various Diameter Pipes
Huafeng Xu, Jinlin Chen
Robotic Intelligence
🎯 What it does: Designed and developed an adaptive pipe inspection robot to accommodate different inner diameters of pipes
Design and Development of a Novel Soft and Inflatable Tactile Sensing Balloon for Early Diagnosis of Colorectal Cancer Polyps
Ozdemir Can Kara, F. Alambeigi
Biomedical Data
🎯 What it does: Proposed and fabricated an inflatable soft visual tactile sensing balloon (VTSB) for assisting colonoscopy in detecting colorectal polyps.
Design and Development of a Rapidly Deployable Low-Cost Tensegrity In-Pipe Robot
Yixiang Liu, Yibin Li
Robotic Intelligence
🎯 What it does: Designed and manufactured a fast, low-cost, insect-shaped tension structure pipeline robot, and conducted experimental validation in various pipeline environments.
Design and Evaluation of Bidirectional Continuous Rotation and Variable Curvature Needle Steering Algorithm
Farid Tavakkolmoghaddam, Gregory S. Fischer
Robotic IntelligenceBiomedical Data
🎯 What it does: Proposed and verified a bidirectional continuous rotation variable curvature needle steering algorithm, achieving variable curvature and closed-loop guidance by controlling the needle's rotational speed;
Design and Implementation of a Two-Limbed 3T1R Haptic Device
Long Kang, Byung-Ju Yi
Robotic Intelligence
🎯 What it does: Proposed a 3T1R tactile device consisting of only two limbs, achieving three translational degrees of freedom and one rotational degree of freedom of motion;
Design and Motion Guidelines for Quadrupedal Locomotion of Maximum Speed or Efficiency with Serial and Parallel Legs
Konstantinos Machairas, E. Papadopoulos
OptimizationRobotic Intelligence
🎯 What it does: By deriving actuator requirements for quadrupedal motion under constant speed and height, a method is proposed to simplify the six-bar three-hop/double-step model to a single-leg model, with analysis conducted on two common two-stage leg architectures (serial and parallel);
Design and Stiffness Analysis of a Bio-Inspired Soft Actuator with Bi-Direction Tunable Stiffness Property
Jianfeng Lin, Zhao Guo
Robotic Intelligence
🎯 What it does: Proposed and implemented a bidirectional tunable stiffness soft actuator (BTSA), integrating air-tendon hybrid drive with bone-shaped structure, and completed its fabrication process and stiffness analysis model;
Design and Synchronous Control of a Magnetically-Actuated and Ultrasound-Guided Multi-Arm Robotic System
Zhengyang Li, Qingsong Xu
Robotic IntelligenceImageUltrasound
🎯 What it does: Designed and verified a multi-arm robot system with mobile magnetic drive and external ultrasound guidance for magnetic catheter manipulation.
Design and Testing of a Flexure-Based XYZ Micropositioner with High Space-Utilization Efficiency
Z. Lyu, Qingsong Xu
Optimization
🎯 What it does: Designed and tested a flexible-base XYZ micro-displacement stage with high spatial utilization
Design for Hip Abduction Assistive Device Based on Relationship Between Hip Joint Motion and Torque During Running
Myunghyun Lee, Seonwoo Kim
Robotic IntelligenceBiomedical Data
🎯 What it does: Designed and prototyped a hip abduction assistive device based on blade springs
Design Method of a Kangaroo Robot with High Power Legs and an Articulated Soft Tail
Shunnosuke Yoshimura, Masayuki Inaba
Robotic Intelligence
🎯 What it does: Proposed a design method for a kangaroo robot with high-power legs and an articulated soft tail
Design of a Cable Driven Wearable Fitness Device for Upper Limb Exercise
J. Park, Minhyung Lee
Robotic IntelligenceBiomedical Data
🎯 What it does: Designed and evaluated a cable-driven wearable fitness device for upper limb exercise, capable of achieving isometric motion trajectories through active resistance control.
Design of a Jumping Control Framework with Heuristic Landing for Bipedal Robots
Jingwen Zhang, D. Hong
Robotic Intelligence
🎯 What it does: Proposes a bipedal robot jumping control framework that utilizes heuristic landing point planning based on momentum feedback during the flight phase, integrated with an improved kinodynamic motion planner, CoM momentum control, and a low-level whole-body dynamics hierarchical controller
Design of Novel Knee Joint Mechanism of Lower-Limb Exoskeleton to Realize Spatial Motion of Human Knee
Man Bok Hong, Seonwoo Kim
OptimizationRobotic Intelligence
🎯 What it does: Designed a novel spherical knee joint mechanism to achieve spatial motion of the human knee, and proposed a corresponding spherical four-bar linkage and optimization method.
Design of STARQ: A Multimodal Quadrupedal Robot for Running, Climbing, and Swimming
Derek A. Vasquez, Jonathan E. Clark
Robotic Intelligence
🎯 What it does: Designed and demonstrated the quadruped robot STARQ capable of walking, climbing, and swimming, achieving multimodal movement.
Design Space Exploration on Efficient and Accurate Human Pose Estimation from Sparse IMU-Sensing
Iris Fürst-Walter, J. Becker
Data SynthesisPose EstimationTime Series
🎯 What it does: Evaluate different numbers and positions of IMU sensors through simulated design space exploration, generate IMU data, and train deep learning models to achieve efficient and accurate human pose estimation.
Design, Characterization and Control of a Whole-body Grasping and Perching (WHOPPEr) Drone
Weijia Tao, Wenlong Zhang
Robotic Intelligence
🎯 What it does: Designed, characterized, and verified a drone that uses the entire机体 as a gripper to achieve rapid and robust circling and payload delivery.
Design, Modeling, and Control of a Low-Cost and Rapid Response Soft-Growing Manipulator for Orchard Operations
Ryan Dorosh, Ming Luo
Robotic IntelligenceAgriculture Related
🎯 What it does: Designed and implemented a low-cost, fast-response soft-growing robotic arm for orchard operations, capable of linear extension up to 1.2 meters, maximum speed of 0.27 μm/s, and terminal load of 1.4 kg.
Detecting Changes in Functional State: A Comparative Analysis Using Wearable Sensors and a Sensorized Tip
Janire Otamendi, A. Zubizarreta
Anomaly DetectionTime SeriesBiomedical Data
🎯 What it does: Comparison of performance between wearable accelerometers and the author's self-developed sensorized assistive walking device in detecting gait abnormalities