TechcraftingAI Robotics

ArXiv Robotics research for Thursday, June 13, 2024. 00:19: Deep Reinforcement Learning-based Quadcopter Controller: A Practical Approach and Experiments 01:48: LLM-Driven Robots Risk Enacting Discrimination, Violence, and Unlawful Actions 03:33: Trajectory Planning for Autonomous Driving in Unstructured Scenarios Based on Graph Neural Network and Numerical Optimization 05:14: OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning 06:39: Human-Robot Interface for Teleoperated Robotized Planetary Sample Collection and Assembly 08:10: Language-Driven Closed-Loop Grasping with Model-Predictive Trajectory Replanning 10:02: Adaptive Actor-Critic Based Optimal Regulation for Drift-Free Uncertain Nonlinear Systems 11:09: Direct Imitation Learning-based Visual Servoing using the Large Projection Formulation 12:17: Beyond the Frontier: Predicting Unseen Walls from Occupancy Grids by Learning from Floor Plans 13:28: AutomaChef: A Physics-informed Demonstration-guided Learning Framework for Granular Material Manipulation 14:34: EHAZOP: A Proof of Concept Ethical Hazard Analysis of an Assistive Robot 15:31: OpenVLA: An Open-Source Vision-Language-Action Model 17:51: Hands-free teleoperation of a nearby manipulator through a virtual body-to-robot link 19:23: Teleoperation of a robotic manipulator in peri-personal space: a virtual wand approach 21:05: RoTipBot: Robotic Handling of Thin and Flexible Objects using Rotatable Tactile Sensors

ArXiv Robotics research for Wednesday, June 12, 2024. 00:19: Hierarchical Reinforcement Learning for Swarm Confrontation with High Uncertainty 01:35: 100 Drivers, 2200 km: A Natural Dataset of Driving Style toward Human-centered Intelligent Driving Systems 03:08: Undergraduate Robotics Education with General Instructors using a Student-Centered Personalized Learning Framework 05:12: Metasensor: a proposal for sensor evolution in robotics 05:55: Highly agile flat swimming robot 06:56: Design, modeling, and characteristics of ringshaped robot actuated by functional fluid 08:15: Chemistry3D: Robotic Interaction Benchmark for Chemistry Experiments 09:49: Learning-based Traversability Costmap for Autonomous Off-road Navigation 10:59: A Hybrid Task-Constrained Motion Planning for Collaborative Robots in Intelligent Remanufacturing 12:41: Review of Autonomous Mobile Robots for the Warehouse Environment 13:50: Trajectory optimization of tail-sitter considering speed constraints 14:55: Utilizing Navigation Path to Generate Target Point for Enhanced End-to-End Autonomous Driving Planning 16:05: PRIBOOT: A New Data-Driven Expert for Improved Driving Simulations 17:41: AToM-Bot: Embodied Fulfillment of Unspoken Human Needs with Affective Theory of Mind

ArXiv Robotics research for Tuesday, June 11, 2024. 00:19: GPU-Accelerated Optimization-Based Collision Avoidance 01:14: Optimal Gait Design for a Soft Quadruped Robot via Multi-fidelity Bayesian Optimization 02:27: Optimal Gait Control for a Tendon-driven Soft Quadruped Robot by Model-based Reinforcement Learning 03:45: A Neck Orthosis with Multi-Directional Variable Stiffness for Persons with Dropped Head Syndrome 04:59: 3D Voxel Maps to 2D Occupancy Maps for Efficient Path Planning for Aerial and Ground Robots 06:31: OTO Planner: An Efficient Only Travelling Once Exploration Planner for Complex and Unknown Environments 07:51: Instruct Large Language Models to Drive like Humans 09:20: Realistic Data Generation for 6D Pose Estimation of Surgical Instruments 11:07: iMESA: Incremental Distributed Optimization for Collaborative Simultaneous Localization and Mapping 12:05: Improving the realism of robotic surgery simulation through injection of learning-based estimated errors 13:33: BAKU: An Efficient Transformer for Multi-Task Policy Learning 14:51: A3VLM: Actionable Articulation-Aware Vision Language Model

ArXiv Robotics research for Monday, June 10, 2024. 00:19: WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts 01:52: Navigation and 3D Surface Reconstruction from Passive Whisker Sensing 02:49: Influence of Motion Restrictions in an Ankle Exoskeleton on Gait Kinematics and Stability in Straight Walking 04:03: Sim-To-Real Transfer for Visual Reinforcement Learning of Deformable Object Manipulation for Robot-Assisted Surgery 05:34: Nonlinear Model Predictive Control of Tiltrotor Quadrotors with Feasible Control Allocation 06:51: Stabilized Adaptive Steering for 3D Sonar Microphone Arrays with IMU Sensor Fusion 07:53: Human Gaze and Head Rotation during Navigation, Exploration and Object Manipulation in Shared Environments with Robots 09:01: A quantitative investigation for deployment of mobile collaborative robots in high-value manufacturing 11:05: Multicam-SLAM: Non-overlapping Multi-camera SLAM for Indirect Visual Localization and Navigation 12:22: An Empirical Design Justice Approach to Identifying Ethical Considerations in the Intersection of Large Language Models and Social Robotics 14:02: Notes on Various Errors and Jacobian Derivations for SLAM 15:49: Notes on Kalman Filter (KF, EKF, ESKF, IEKF, IESKF) 16:45: Towards Real-World Efficiency: Domain Randomization in Reinforcement Learning for Pre-Capture of Free-Floating Moving Targets by Autonomous Robots 18:01: Demonstrating HumanTHOR: A Simulation Platform and Benchmark for Human-Robot Collaboration in a Shared Workspace

ArXiv Robotics research for Sunday, June 09, 2024. 00:19: A Superalignment Framework in Autonomous Driving with Large Language Models 02:03: FlightBench: A Comprehensive Benchmark of Spatial Planning Methods for Quadrotors 03:21: Towards A General-Purpose Motion Planning for Autonomous Vehicles Using Fluid Dynamics 05:14: TR2MTL: LLM based framework for Metric Temporal Logic Formalization of Traffic Rules 06:35: Fast and Certifiable Trajectory Optimization 08:16: MAP-ADAPT: Real-Time Quality-Adaptive Semantic 3D Maps 09:38: Differentiable Discrete Elastic Rods for Real-Time Modeling of Deformable Linear Objects 10:59: Open-Vocabulary Part-Based Grasping 12:15: Visual-Inertial SLAM as Simple as A, B, VINS 13:24: LOP-Field: Brain-inspired Layout-Object-Position Fields for Robotic Scene Understanding

ArXiv Robotics research for Saturday, June 08, 2024. 00:19: Metaverse for Safer Roadways: An Immersive Digital Twin Framework for Exploring Human-Autonomy Coexistence in Urban Transportation Systems 01:47: A preprocessing-based planning framework for utilizing contacts in high-precision insertion tasks 03:20: Trust the PRoC3S: Solving Long-Horizon Robotics Problems with LLMs and Constraint Satisfaction 04:48: Toward Autonomous Driving by Musculoskeletal Humanoids: A Study of Developed Hardware and Learning-Based Software 05:52: A Survey on Hybrid Motion Planning Methods for Automated Driving Systems 07:33: Distributed Motion Control of Multiple Mobile Manipulator System with Disturbance and Communication Delay 08:42: Cross Language Soccer Framework: An Open Source Framework for the RoboCup 2D Soccer Simulation 09:58: Observation Denoising in CYRUS Soccer Simulation 2D Team For RoboCup 2024

ArXiv Robotics research for Friday, June 07, 2024. 00:19: A Hybrid-Layered System for Image-Guided Navigation and Robot Assisted Spine Surgery 01:58: Underactuated Control of Multiple Soft Pneumatic Actuators via Stable Inversion 03:03: Scaling Motion Planning Infeasibility Proofs 04:05: TEDi Policy: Temporally Entangled Diffusion for Robotic Control 04:57: Optimal path planning and weighted control of a four-arm robot in on-orbit servicing 06:11: Deep Learning Powered Estimate of The Extrinsic Parameters on Unmanned Surface Vehicles 07:14: SLR: Learning Quadruped Locomotion without Privileged Information 08:24: Auto-Multilift: Distributed Learning and Control for Cooperative Load Transportation With Quadrotors 10:01: InstructNav: Zero-shot System for Generic Instruction Navigation in Unexplored Environment 11:35: A Modular Framework for Flexible Planning in Human-Robot Collaboration 12:33: Sim-to-real Transfer of Deep Reinforcement Learning Agents for Online Coverage Path Planning 13:57: Robotic in-hand manipulation with relaxed optimization 15:09: Experimental Evaluation of ROS-Causal in Real-World Human-Robot Spatial Interaction Scenarios 16:45: Designs for Enabling Collaboration in Human-Machine Teaming via Interactive and Explainable Systems 18:23: I2EDL: Interactive Instruction Error Detection and Localization

ArXiv Robotics research for Thursday, June 06, 2024. 00:19: Phase-Amplitude Reduction-Based Imitation Learning 01:28: RoboCoder: Robotic Learning from Basic Skills to General Tasks with Large Language Models 02:54: Touch100k: A Large-Scale Touch-Language-Vision Dataset for Touch-Centric Multimodal Representation 04:43: Bench2Drive: Towards Multi-Ability Benchmarking of Closed-Loop End-To-End Autonomous Driving 06:48: A Survey of Language-Based Communication in Robotics 08:14: Data-driven Explainable Controller for Soft Robots based on Recurrent Neural Networks 09:45: Redundancy-aware Action Spaces for Robot Learning 10:58: MARLander: A Local Path Planning for Drone Swarms using Multiagent Deep Reinforcement Learning 12:12: Text-to-Drive: Diverse Driving Behavior Synthesis via Large Language Models

ArXiv Robotics research for Wednesday, June 05, 2024. 00:19: Real-time Motion Planning for autonomous vehicles in dynamic environments 01:48: Robots Have Been Seen and Not Heard: Effects of Consequential Sounds on Human-Perception of Robots 03:21: FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality 04:49: Learning Semantic Traversability with Egocentric Video and Automated Annotation Strategy 06:27: Balancing Performance and Efficiency in Zero-shot Robotic Navigation 07:24: Adaptive Distance Functions via Kelvin Transformation 08:29: Object Manipulation in Marine Environments using Reinforcement Learning 09:49: Mission Design for Unmanned Aerial Vehicles using Hybrid Probabilistic Logic Program

ArXiv Robotics research for Tuesday, June 04, 2024. 00:19: Improving Generalization in Aerial and Terrestrial Mobile Robots Control Through Delayed Policy Learning 01:27: Exploring Real World Map Change Generalization of Prior-Informed HD Map Prediction Models 02:45: DrEureka: Language Model Guided Sim-To-Real Transfer 03:57: Cross-Embodiment Robot Manipulation Skill Transfer using Latent Space Alignment 05:31: Kinematic analysis of a parallel robot for minimally invasive surgery 06:19: MS-Mapping: Multi-session LiDAR Mapping with Wasserstein-based Keyframe Selection 07:37: An Observability-Constrained Magnetic-Field-Aided Inertial Navigation System 08:21: A Multipurpose Interface for Close- and Far-Proximity Control of Mobile Collaborative Robots 10:16: An Open and Reconfigurable User Interface to Manage Complex ROS-based Robotic Systems 11:48: Broadcasting Support Relations Recursively from Local Dynamics for Object Retrieval in Clutters 13:04: Train Localization During GNSS Outages: A Minimalist Approach Using Track Geometry And IMU Sensor Data 14:08: Exploiting Chordal Sparsity for Fast Global Optimality with Application to Localization 15:12: Query-based Semantic Gaussian Field for Scene Representation in Reinforcement Learning 16:48: Out-of-Distribution Runtime Adaptation with Conformalized Neural Network Ensembles 18:19: RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots

ArXiv Robotics research for Monday, June 03, 2024. 00:19: Multi-Object Tracking based on Imaging Radar 3D Object Detection 01:42: Hybrid Quadratic Programming -- Pullback Bundle Dynamical Systems Control 03:12: Ubiquitous Robot Control Through Multimodal Motion Capture Using Smartwatch and Smartphone Data 04:08: Configuration Space Distance Fields for Manipulation Planning 06:02: Learning-based legged locomotion; state of the art and future perspectives 07:04: C$^3$P-VoxelMap: Compact, Cumulative and Coalescible Probabilistic Voxel Mapping 08:47: Deep Stochastic Kinematic Models for Probabilistic Motion Forecasting in Traffic 10:39: Learning from Mistakes: a Weakly-supervised Method for Mitigating the Distribution Shift in Autonomous Vehicle Planning 12:17: ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation 13:38: PlanAgent: A Multi-modal Large Language Agent for Closed-loop Vehicle Motion Planning

ArXiv Robotics research for Sunday, June 02, 2024. 00:19: An Efficient Trajectory Generation for Bi-copter Flight in Tight Space 01:31: MINER-RRT*: A Hierarchical and Fast Trajectory Planning Framework in 3D Cluttered Environments 02:47: QUADFormer: Learning-based Detection of Cyber Attacks in Quadrotor UAVs 04:16: Unsupervised Neural Motion Retargeting for Humanoid Teleoperation 05:31: Accelerate Hybrid Model Predictive Control using Generalized Benders Decomposition 06:40: Arena 3.0: Advancing Social Navigation in Collaborative and Highly Dynamic Environments 08:21: Understanding On-the-Fly End-User Robot Programming 09:22: Using 3-D LiDAR Data for Safe Physical Human-Robot Interaction 10:24: Deep Reinforcement Learning for Sim-to-Real Policy Transfer of VTOL-UAVs Offshore Docking Operations 12:19: Efficient Behavior Tree Planning with Commonsense Pruning and Heuristic 13:56: Evaluating MEDIRL: A Replication and Ablation Study of Maximum Entropy Deep Inverse Reinforcement Learning for Human Social Navigation

ArXiv Robotics research for Saturday, June 01, 2024. 00:19: NuRF: Nudging the Particle Filter in Radiance Fields for Robot Visual Localization 01:23: Precision and Adaptability of YOLOv5 and YOLOv8 in Dynamic Robotic Environments 02:36: Cognitive Manipulation: Semi-supervised Visual Representation and Classroom-to-real Reinforcement Learning for Assembly in Semi-structured Environments 04:12: Teledrive: An Embodied AI based Telepresence System 05:38: Evaluating Uncertainty-based Failure Detection for Closed-Loop LLM Planners 07:23: Learning Manipulation by Predicting Interaction 08:57: Task Planning for Object Rearrangement in Multi-room Environments 10:24: TacShade A New 3D-printed Soft Optical Tactile Sensor Based on Light, Shadow and Greyscale for Shape Reconstruction 11:56: Research on an Autonomous UAV Search and Rescue System Based on the Improved 12:59: Learning to Play Air Hockey with Model-Based Deep Reinforcement Learning

ArXiv Robotics summaries for May 31, 2024

ArXiv Robotics summaries for May 30, 2024

ArXiv Robotics summaries for May 29, 2024

ArXiv Robotics summaries for May 28, 2024

ArXiv Robotics summaries for May 27, 2024

ArXiv Robotics summaries for May 26, 2024

ArXiv Robotics summaries for May 25, 2024

ArXiv Robotics summaries for May 24, 2024

ArXiv Robotics summaries for May 23, 2024

arXiv Robotics research summaries for May 22, 2024. Today's Research Themes (AI-Generated): • Novel algorithm enables autonomous vehicles to learn and improve driving behaviors with minimal human resets during reinforcement learning. • Enhanced mid-air drone interception system utilizing novel guidance and estimation methods surpasses existing solutions. • HighwayLLM integrates large language models with reinforcement learning for interpretable decision-making in highway autonomous driving. • Robot chain-of-thought approach with sparse failure guidance outperforms traditional methods in learning complex manipulation tasks. • Waverider system combines hierarchical maps and Riemannian Motion Policies for efficient and reactive obstacle avoidance in mobile robots.

arXiv Computer Vision research summaries for May 21, 2024. Today's Research Themes (AI-Generated): • Introduction of 3DSS-Mamba for efficient hyperspectral image classification with global modeling capabilities. • Development of few-shot learning mechanism for customized image editing with minimal data. • CLRKDNet balances lane detection accuracy and real-time performance for autonomous driving. • NOVA-3D framework enables direct reconstruction of 3D anime characters from 2D non-overlapped views. • New knowledge aggregation and distillation method improves active object detection in complex scenes.

arXiv Robotics research summaries for May 20, 2024. Today's Research Themes (AI-Generated): • CDM-MPC offers a novel framework for dynamic bipedal robot jumping, ensuring stability via motion planning and real-time control. • An active learning approach is demonstrated for real-time haptic training, enhancing tactile perceptual models using biomimetic sensors. • Techniques for self-supervised learning enable low-rigidity robots to adapt visual servoing to temporal body changes for accurate grasping. • Salience-guided ground features improve robot localization in complex urban environments, outperforming traditional geometric methods. • Octo, an open-source generalist robot policy, shows promise in leveraging diverse sensor inputs and action spaces for robotic manipulation.

arXiv Computer Vision research summaries for May 19, 2024. Today's Research Themes (AI-Generated): • PQ3D demonstrates a unified model for 3D vision-language tasks with impressive multi-task training performance, setting new benchmarks. • AdaAugment, a novel data augmentation method, utilizes reinforcement learning for dynamic adjustment, outperforming state-of-the-art methods. • New cross-domain knowledge distillation framework boosts low-resolution human pose estimation, introducing scale-adaptive and cross-class modules. • Era3D, using row-wise attention, significantly enhances multiview diffusion for high-resolution image generation from single-view inputs. • Tangle leverages gene expression profiles for unsupervised slide representation learning, showing superior few-shot performance on pathology datasets.

arXiv Robotics research summaries for May 18, 2024. Today's Research Themes (AI-Generated): • Ground segmentation and outlier-robust methodologies improve long-term robotic mapping for diverse real-world scenarios. • PS6D demonstrates significant improvements in 6D object pose estimation for industrial applications, particularly in challenging bin-picking tasks. • Visual episodic memory enhances robotic exploration by using intrinsic motivation signals derived from spatiotemporal feature predictions. • A safety framework combining temporal logic and reachability analysis provides automated safety guarantees for intelligent vehicle intersections. • Meta-Control facilitates customizable state representation and control strategies for robots to handle heterogeneous tasks with varied requirements.

arXiv Robotics research summaries for May 17, 2024. Today's Research Themes (AI-Generated): • Advancements in autonomous vehicle assessment through complex multi-agent testing scenarios in Singapore traffic. • Efficient reinforcement learning control framework for artificial muscle-driven robotics with sim-to-real transfer. • Novel occupancy grid map-based control barrier functions for safe navigation in unknown environments. • Simultaneous optimization of robot trajectory and continuous occupancy maps with new SLAM approach. • Establishing safer automated driving with fast collision probability estimation using multi-circular shape approximations.

arXiv Robotics research summaries for May 16, 2024. Today's Research Themes (AI-Generated): • SEEK framework combines semantic knowledge with robotic observations for more efficient object-goal navigation in inspections. • Nonlinear Model Predictive Control in overactuated tiltable-quadrotors enables rapid and robust pose tracking. • ACES introduces a teleoperated robotic solution for internal non-destructive inspection of concrete pipes. • Natural language descriptions bridge the sim2real gap by providing a domain-invariant image representation for robotic policies. • Multi Q-learning optimizes UAV connectivity for search and rescue operations in challenging terrains.

arXiv Robotics research summaries for May 15, 2024. Today's Research Themes (AI-Generated): • Explainable AI for ship collision avoidance visualizes decision-making and behavioral intentions to improve safety and interpretability. • Adaptive Koopman architecture enhances robust control for dynamic systems, adaptable to changes and disturbances. • Gaussian process models for predicting human motion enhance safety and efficiency in collaborative robot-human interactions in virtual environments. • CarDreamer is an open-source world model-based platform improving autonomous vehicle training with extensive configurable tasks and visualization tools. • BonnBot-I Plus advances precision weeding in agriculture, integrating biodiversity considerations and presenting real-world robotic platform performance.

arXiv Robotics research summaries for May 14, 2024. Today's Research Themes (AI-Generated): • Cross-category functional grasp synthesis improves object manipulation by leveraging a knowledge graph of similarities in shapes and grasp types. • Analytic vehicle dynamics model for real-world maneuver planning combines speed with real-world accuracy under simplifying assumptions. • Efficient multi-robot rendezvous in unknown environments is achieved through partitioned exploration and communication-efficient topological mapping. • Real-time global optimization enables fail-safe emergency stops in automated driving, despite electrical or electronic failures. • The IPC method offers robust performance in SLAM backends by incrementally finding a consistent set of measurements amidst outliers.

arXiv Computer Vision research summaries for May 13, 2024. Today's Research Themes (AI-Generated): • DualFocus enhances text-based person retrieval with integrated positive and negative descriptors for more accurate vision-language matching. • GaussianVTON revolutionizes 3D virtual try-on using multi-stage Gaussian Splatting editing with image prompting for e-commerce applications. • Text Grouping Adapter adapts pre-trained text detectors for efficient layout analysis, improving context capture for text grouping. • Support-Query Prototype Fusion Network advances few-shot medical image segmentation with superior support-query fused prototype construction. • Deep learning, prior-based, and hybrid approaches in dehazing remote sensing and UAV imagery are exhaustively reviewed in the context of contemporary challenges and future research directions.

arXiv Robotics research summaries for May 12, 2024. Today's Research Themes (AI-Generated): • Integration of Large Language Models improves robot reward learning without human input, enhancing training efficacy and efficiency. • Air-ground robot teams employ semantics for large-scale exploration, overcoming challenges in hazardous and unstructured environments with open-source code. • Vision-Based Tactile Sensors facilitate realistic simulation and successful manipulation of deformable objects with a high sim-to-real transfer success rate. • BeautyMap leverages a binary-encoded matrix for dynamic point removal, improving accuracy and efficiency in global map generation for static environments. • DiffGen framework leverages differentiable physics and rendering, plus vision-language models, for efficient, minimal-effort robot demonstration generation.

arXiv Robotics research summaries for May 11, 2024. Today's Research Themes (AI-Generated): • A novel human-in-the-loop approach optimizes robot-assisted feeding by balancing task performance and cognitive workload. • Autonomous harvesting robot AHPPEBot demonstrates high success rates in commercial greenhouses, advancing agricultural automation. • ManiFoundation Model introduces a large-scale manipulator model for general-purpose robotic handling, achieving high success rates across diverse objects. • Mobius transformation-based control enables unicycle robots to maintain stability within complex nonconcentric circular boundaries. • Multi-agent Traffic Prediction model innovates in high-speed environments by incorporating agents' intentions and uncertainties.

arXiv Robotics research summaries for May 10, 2024. Today's Research Themes (AI-Generated): • Enhancing off-road semantic mapping using Dempster-Shafer theory to improve uncertainty reliability and semantic accuracy. • Advancing virtual testing for automated driving with the AVEAS project's methods for real-world data acquisition and online database development. • Integrating motion planning and control for precise positioning of car-like robots using continuous-curvature paths and nonlinear model predictive control. • Proposing a novel method for learning diverse 6-DoF grasps for robotic bin picking, outperforming current approaches in efficiency and generalization. • Developing emotion and context-aware behavior trees for ethical robotic decision-making focused on safety, context, privacy, and transparency.

arXiv Robotics research summaries for May 09, 2024. Today's Research Themes (AI-Generated): • Introduction of Dynamic Deep Factor Graphs (DDFG) to enhance multi-agent reinforcement learning with adaptable value function structures. • Development of a universal LiDAR calibration method leveraging a coarse-to-fine framework for autonomous systems. • ASGrasp presents a network for grasping transparent objects using RGB-D active stereo camera, achieving high success rates. • LOP-UKF combines LiDAR Odometry and the Pacejka tire model for accurate real-time racecar sideslip estimation. • Proposal of a passive control policy that adapts to obstacles for safe, robust autonomous robot motion.

arXiv Robotics research summaries for May 08, 2024. Today's Research Themes (AI-Generated): • Explore latent codes as an alternative to language for hierarchical robotic control that outperforms GPT-4V in complex tasks. • Delve into the challenges and advancements in place recognition for enhancing real-world autonomy in robotics. • Discover the Aerial Tendon-Driven Manipulator's innovative design for improved stiffness and reduced aerial manipulator disturbances. • Investigate novel safety-critical compliant control strategies for robots interacting with unknown environments while maintaining force constraints. • Assess the effectiveness of robotics localization methods utilizing prior topological and metric information in a graph-based framework.

arXiv Robotics research summaries for May 07, 2024. Today's Research Themes (AI-Generated): • DInEKF improves multi-robot cooperative localization with matrix Lie groups for better accuracy and consistency in 3-D environments. • Stakeholder-specific pruning methodologies pave the way for robotic automation in agriculture. • Enhancements in event-based visual odometry are achieved through IMU integration and efficient edge-pixel sampling strategies. • Robot swarms are proposed as supportive systems within heterogeneous groups, enhancing data acquisition in large unknown environments. • Logic-Skill Programming combines independently learned robot skills to optimize complex task sequences through symbolic search and value optimization.

arXiv Robotics research summaries for May 06, 2024. Today's Research Themes (AI-Generated): • CushSense introduces cost-effective, soft tactile-sensing skin for safe human-robot interaction, emphasizing comfort in caregiving applications. • Predictive uncertainty in robotic systems is crucial for real-world applications, with a guide provided for quantifying and managing it. • A novel framework for evaluating drivers’ interaction abilities is proposed to improve autonomous vehicle adaptability in social driving scenarios. • Fast and Direct Smooth Path Planning method developed to create smooth robot paths with constant velocity, outperforming traditional algorithms in multiple scenarios. • Review of AI in autonomous endovascular interventions underscores the experimental stage of technology and highlights challenges for clinical application.

arXiv Robotics research summaries for May 05, 2024. Today's Research Themes (AI-Generated): • Satellite onboard computing holds potential for real-time road flooding detection to improve navigation during natural disasters. • The DexiTac soft gripper exemplifies a reconfigurable approach to robot manipulation combining soft dexterity with tactile sensing. • A simulation framework for optical tactile sensors enhances robotic perception of complex manipulations like slip and rotation. • FreeAlign enables robust collaborative perception for multiple agents without reliance on external localization and clock devices. • CoverLib introduces classifiers-equipped experience libraries to optimize domain-tuned motion planning for robotics.

arXiv Robotics research summaries for May 04, 2024. Today's Research Themes (AI-Generated): • Investigating the cognitive benefits of robotic therapies for elderly individuals with mild cognitive impairments. • Advancing swarm robot behaviors by minimizing surprise through neuroevolution to enhance robustness and scalability. • Proposing an efficient approximation for information gain to improve decision-making in autonomous robotics. • Discovering expert racing drivers' cognitive and adaptive strategies to inform the development of autonomous racing algorithms. • Enhancing wall-climbing robot stability with a gravity compensation strategy inspired by gecko locomotion.

arXiv Robotics research summaries for May 03, 2024. Today's Research Themes (AI-Generated): • RobHAR advances human activity recognition with a robot-mounted mmWave radar for real-time elderly care monitoring. • A new onboard framework enhances staircase detection and modeling for legged robots using point cloud data and data augmentation techniques. • Task-driven computational framework proposed for optimizing the design and mounted pose of modular reconfigurable manipulators. • Development of a tendon-driven variable stiffness manipulator demonstrates increased payload capacity and improved stability. • Sonar-based positioning system designed for AUVs offers robust localization in underwater environments with sparse infrastructure.

arXiv Robotics research summaries for May 02, 2024. Today's Research Themes (AI-Generated): • Advances in empathic robotics explore the relationship between perceived empathy and mental state attribution in human-robot interactions. • Differentiable Particles (DiPac) demonstrates a general-purpose algorithm for deformable object manipulation across various object types. • Novel continual learning techniques enable robust gate detection in autonomous drone racing, even under dynamic lighting conditions. • CoViS-Net enhances multi-robot cooperative tasks through a decentralized visual spatial foundation model that predicts without camera overlap. • A new self-alignment method for SINS in autonomous vehicles allows simultaneous latitude and attitude determination without external aid.

arXiv Robotics research summaries for May 01, 2024. Today's Research Themes (AI-Generated): • FusionAssurance framework enhances autonomous driving with physics-informed control for improved safety and generalization. • Implicit SVSDF contributes to continuous collision avoidance in trajectory generation for objects of varied shapes and dynamics. • Tactile Reinforcement Learning shows promise for robotic insertion tasks, leveraging tactile feedback for improved manipulation performance. • Ultrasound Embodied Intelligence system integrates LLM with domain knowledge for efficient, autonomous medical ultrasound scanning. • GAD framework aims to surpass rule-based methods in HD map-free autonomous driving by integrating prediction, decision, and planning.

arXiv Robotics research summaries for April 30, 2024. Today's Research Themes (AI-Generated): • Improving human-robot collaboration with reinforcement learning for nonverbal auditory expressions. • Introducing DiffuseLoco, a real-time control framework for dynamic legged locomotion utilizing diffusion models. • Enhancing urban robot navigation with audio-visual traffic light state detection overcoming visual occlusions. • Presenting a user interface for explainable AI-based optimization in industrial robotics for varied skill levels. • Developing Transformer-Enhanced Motion Planner to refine sampling-based motion planning via attention mechanisms.

arXiv NLP research summaries for April 29, 2024. Today's Research Themes (AI-Generated): • Med-Gemini models set new standards in medical AI by surpassing GPT-4 and demonstrating potential for real-world healthcare applications. • BMRetriever showcases efficient biomedical text retrieval with fewer parameters, outperforming larger models. • Ethical reasoning in large language models is influenced by the language of interaction, with GPT-4 showing the least bias across languages. • Half Fine-Tuning (HFT) method introduced to effectively retain knowledge and mitigate forgetting in large language models. • Evaluating forensic linguistics' AI models for transparency reveals that explainability in machine learning is fundamental for legal applications.

arXiv Robotics research summaries for April 28, 2024. Today's Research Themes (AI-Generated): • Leveraging large vision models for robust, online, target-free LiDAR-camera extrinsic calibration in diverse environments. • Novel VLP/INS integration approach achieves high indoor positioning accuracy with inclination estimation and blockage handling. • Dexterous Grasp TRansformer revolutionizes object grasping with a transformer-based model that predicts diverse and high-quality grasp poses. • Innovative contactless object manipulation using fluid streams demonstrated in a differentiable simulation environment. • ROS 2 hardware implementation greatly enhances speed and efficiency in robotic communication networks.

arXiv Robotics research summaries for April 27, 2024. Today's Research Themes (AI-Generated): • HIPer introduces a human-inspired model for robotic scene perception integrating neuroscience concepts for enhanced decision-making in mobile robots. • Efficient bi-manipulation using RGBD multi-model fusion demonstrated improved performance in intelligent manufacturing with focal attention mechanisms. • Hierarchical learning-based robotic control emulates animal central nervous system structures for advanced legged robot locomotion on uneven terrains. • An adaptive motion planning approach for off-road autonomous vehicles mimics human cognition for trajectory optimization in complex terrains. • Visual Imitation Learning with Waypoints (VIEW) significantly boosts the efficiency of robots learning tasks from video by leveraging condensed trajectory waypoints.

arXiv Robotics research summaries for April 26, 2024. Today's Research Themes (AI-Generated): • Lifelong policy learning enhances autonomous driving with incremental knowledge acquisition and robust policy performance improvement. • Advances in SLAM technology enable detailed indoor mapping for large-scale environments like factory halls and construction sites. • Scene extrapolation generates diverse traffic scenarios for testing highly automated driving functions within simulations. • mmWave radar point clouds are bolstered by visual-inertial supervision, allowing for enhanced object detection and mapping. • A new distance measure reliant on semantic features improves clustering of motion trajectories, benefiting human-robot interaction.

arXiv Robotics research summaries for April 25, 2024. Today's Research Themes (AI-Generated): • GPU-accelerated Stein particle filtering enhances 6-DoF Monte Carlo localization with real-time robustness against sensor occlusion. • SwarmRL offers a comprehensive toolkit for research on intelligent active particle systems using classical control and deep reinforcement learning. • Few-shot imitation learning on robotic hands is improved using pretrained latent representations for enhanced error resilience and manipulation. • Conformalized prediction models enable motion controllers of automated vehicles to adjust for actuator degradations and maintain safe operation. • Multifingered hands equipped with visuotactile data enable learning of complex bimanual manipulations for more human-like dexterity and skill.