Robotics & Automation
Robotics & Automation news — Monday 19 October 2026
130 stories from 4 sources, every one in English. Each links to its publisher.
-
Unleash AI Innovation: The Power of NVIDIA RTX PRO 6000 Blackwell Workstation Edition Fueled by PNY-Supplied GPUs
Robotics Business Review
By PNY Technologies Experiment, Prototype, and Innovate with AI Artificial intelligence is transforming industries at an unprecedented pace. From agentic AI that autonomously solves complex, multi-step problems to physical AI that brings…
-
Unleash AI Innovation: The Power of NVIDIA RTX PRO 6000 Blackwell Workstation Edition Fueled by PNY-Supplied GPUs
The Robot Report
By PNY Technologies Experiment, Prototype, and Innovate with AI Artificial intelligence is transforming industries at an unprecedented pace. From agentic AI that autonomously solves complex, multi-step problems to physical AI that brings…
-
Daimon Robotics building the infrastructure layer for tactile intelligence
The Robot Report
Credit: Daimon Robotics Vision has become central to modern robotics, but cameras alone cannot tell a robot everything it needs to know about physical contact. When handling flexible objects, inserting components, or manipulating delicate…
-
Daimon Robotics building the infrastructure layer for tactile intelligence
Robotics Business Review
Credit: Daimon Robotics Vision has become central to modern robotics, but cameras alone cannot tell a robot everything it needs to know about physical contact. When handling flexible objects, inserting components, or manipulating delicate…
-
Kodama raises $7M to bring robotics to forestry
The Robot Report
Kodama initially focused on wildfire risk reduction in the western U.S., but it is now focusing on the timber industry in the southeastern U.S. | Source: Kodama Kodama Systems today said it has raised $7 million to bring physical AI to…
-
Kodama raises $7M to bring robotics to forestry
Robotics Business Review
Kodama initially focused on wildfire risk reduction in the western U.S., but it is now focusing on the timber industry in the southeastern U.S. | Source: Kodama Kodama Systems today said it has raised $7 million to bring physical AI to…
-
The Robot I Didn’t Expect to Build
IEEE Spectrum Robotics
Agility Robotics , the company I cofounded out of my robotics lab at Oregon State University in 2015, has just launched Digit 5 . It’s designed to be the first humanoid robot that can safely and economically do useful work at scale…
-
A guide to everything happening at RoboBusiness 2026
The Robot Report
RoboBusiness 2026 kicks off tomorrow at the Santa Clara Convention Center. Over the next two days, the show will bring together over 1,600 attendees for discussions on cutting-edge research, industry trends, and innovative applications in…
-
A guide to everything happening at RoboBusiness 2026
Robotics Business Review
RoboBusiness 2026 kicks off tomorrow at the Santa Clara Convention Center. Over the next two days, the show will bring together over 1,600 attendees for discussions on cutting-edge research, industry trends, and innovative applications in…
-
TriSAR-3D and PaintFusion: A Novel Trimodal LiDAR-RGB-Thermal Dataset and Method for 3D Person Detection in Search and Rescue
arXiv cs.RO (Robotics)
arXiv:2610.22509v1 Announce Type: cross Abstract: Reliable 3D person detection is important for rescue robotics to give first responders situational awareness. Yet most multimodal 3D detection research targets automotive scenarios in which…
-
StopWatch: Goal Grounding for Object-goal Navigation
arXiv cs.RO (Robotics)
arXiv:2610.23841v1 Announce Type: cross Abstract: Pretrained VLMs bring rich world knowledge and semantic priors to object-goal navigation, and fine-tuning a VLM into a navigation policy has become a common recipe. Despite this progress…
-
TRAIN: Terrain-Aware Interactive Navigation for Quadruped Robots via Reinforcement Learning
arXiv cs.RO (Robotics)
arXiv:2602.12724v4 Announce Type: replace Abstract: This study introduces TRAIN: TeRrain-Aware Interactive Navigation, a deep reinforcement learning (DRL)-based framework for quadrupedal navigation across unstructured terrain and…
-
OpenDexGrasp: Open-vocabulary Task-Oriented Dexterous Grasping
arXiv cs.RO (Robotics)
arXiv:2609.18117v3 Announce Type: replace Abstract: Dexterous grasp synthesis has advanced rapidly in generating stable and physically plausible hand poses, but real-world manipulation requires grasps that preserve the function implied by…
-
RoboHarness: A Simple Harness Outperforms VLA and World Action Models
arXiv cs.RO (Robotics)
arXiv:2610.14018v2 Announce Type: replace Abstract: The rise of large language models (LLMs) has opened a path toward generalist embodied AI, raising the question of how LLMs can act in the physical world. One straightforward approach is…
-
AnyRank: Reliable Anytime Policy Ranking via Sequential Testing
arXiv cs.RO (Robotics)
arXiv:2610.14603v2 Announce Type: replace Abstract: Due to the high cost of hardware experiments, robot policies are predominantly evaluated by comparing their empirical performance on a limited number of hardware trials. However…
-
Metro-WM: Long-Horizon Latent Planning with Realisable Sub-Goals
arXiv cs.RO (Robotics)
arXiv:2609.31868v2 Announce Type: replace-cross Abstract: Model-predictive control with Joint-Embedding Predictive Architectures (JEPAs) provides a strong zero-shot goal-reaching planner, but it is only effective over short planning…
-
GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed
arXiv cs.RO (Robotics)
arXiv:2609.39600v2 Announce Type: replace-cross Abstract: Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence…
-
RealtimeWAM: One-Step Asynchronous World Action Models
arXiv cs.RO (Robotics)
arXiv:2610.06617v2 Announce Type: replace-cross Abstract: World Action Models (WAMs) incorporate visual representations from video generation backbones to guide action prediction. Recent efficient WAMs adopt Mixture-of-Transformers (MoT)…
-
SidewalkBench: Benchmarking Visual Navigation on Urban Sidewalks
arXiv cs.RO (Robotics)
arXiv:2606.16953v3 Announce Type: replace Abstract: Urban sidewalk navigation presents significant challenges due to complex structural layouts, dynamic pedestrian behaviors, and long distances. While recent visual navigation models offer…
-
Mothership: A Two Tier Autonomous Refuse Collection System for Pedestrian Campuses
arXiv cs.RO (Robotics)
arXiv:2610.22461v1 Announce Type: new Abstract: Refuse collection in pedestrian campuses, parks and mixed-use districts remains manual because compaction trucks cannot reach bins located on walkways, courtyards and building service doors…
-
Has GPT-6-Astra Achieved Embodied Intelligence?
arXiv cs.RO (Robotics)
arXiv:2610.22471v1 Announce Type: new Abstract: GPT-6-Astra demonstrates a level of competence in robotic manipulation and navigation that challenges the conventional paradigm of building specialized models for individual tasks and…
-
Pedicel-Targeted Strawberry Harvesting via Imitation Learning on a Low-Cost Teleoperated Robotic Arm
arXiv cs.RO (Robotics)
arXiv:2610.22765v1 Announce Type: new Abstract: Precise pedicel-targeted grasping in strawberry harvesting remains a critical bottleneck due to the pedicel's small diameter (1.3-3 mm), color similarity with surrounding foliage, and precise…
-
SCOPE: Selective Certification of Program Edits for Embodied Heuristic Learning
arXiv cs.RO (Robotics)
arXiv:2610.23128v1 Announce Type: new Abstract: Embodied heuristic learning must balance strategic adaptation with the preservation of previously successful behavior. Local repairs may leave structural failures unresolved, whereas broader…
-
A Tendon-Driven Robotic Jellyfish with Constrained Soft Actuation and Depth Control via Reinforcement Learning
arXiv cs.RO (Robotics)
arXiv:2609.29132v2 Announce Type: replace Abstract: Jellyfish-inspired robots offer a compliant and efficient approach to underwater locomotion, but achieving large deformation together with repeatable actuation and closed-loop control…
-
Confidence Calibration in Vision-Language-Action Models
arXiv cs.RO (Robotics)
arXiv:2507.17383v3 Announce Type: replace Abstract: Trustworthy robot behavior requires not only high levels of task success but also that the robot can reliably quantify how likely it is to succeed. To this end, we present a…
-
Embodiment-Aware Planning for Heterogeneous Bimanual Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.23551v1 Announce Type: new Abstract: Many manipulation tasks that are effortless for dexterous human hands remain challenging for robotic systems equipped with simple, task-specialized grippers. Industrial platforms commonly…
-
PinNet: Keypoint-Aware Learned Local Descriptors with Geometric Embedding for Loop Closure in LiDAR SLAM
arXiv cs.RO (Robotics)
arXiv:2606.28637v3 Announce Type: replace Abstract: Loop closure is essential to reduce drift and build globally consistent maps in large-scale environments. However, reliable loop closure with only geometric information from, e.g., a…
-
Towards Robot Skill Learning and Adaptation with Gaussian Processes
arXiv cs.RO (Robotics)
arXiv:2603.01480v2 Announce Type: replace Abstract: General robot skill adaptation requires expressive representations robust to varying task configurations. While recent learning-based skill adaptation methods refined via Reinforcement…
-
Learning Cross-Coupled and Regime Dependent Dynamics for Aerial Manipulation
arXiv cs.RO (Robotics)
arXiv:2605.14805v2 Announce Type: replace Abstract: Accurate dynamics modeling of aerial manipulators is challenging because platform--manipulator coupling, history-dependent interactions, and changing payload or manipulator configurations…
-
Learning Asynchronous Upper-body Task-space Trajectory Tracking Policy for Humanoid Robots
arXiv cs.RO (Robotics)
arXiv:2606.25706v2 Announce Type: replace Abstract: High-level humanoid planners often output sparse task-space, low-rate trajectories, whereas whole-body controllers run at high frequency. This creates temporal asynchrony between the…
-
EDAR: Learning Environment-Dependent Action Representations for Robotic Manipulation
arXiv cs.RO (Robotics)
arXiv:2607.11427v2 Announce Type: replace Abstract: Learning effective action representations is critical for robotic manipulation, where raw control trajectories are often noisy, redundant, and difficult to model directly. Existing…
-
Adapting Generalist Vehicle Models for High-Speed MPC Across Terrains
arXiv cs.RO (Robotics)
arXiv:2607.13319v4 Announce Type: replace Abstract: High-speed off-road autonomy requires precise closed-loop control for a target vehicle while remaining robust across changing terrains. Recent forward kinodynamic (FKD) prediction…
-
Demystifying When and Why VLAs Fail in Contact-Rich Tasks and How to Fix Them
arXiv cs.RO (Robotics)
arXiv:2608.01402v2 Announce Type: replace Abstract: We address the problem of understanding when and why Vision-Language-Action models struggle with contact-rich manipulation tasks that require precise physical interaction. Prior work has…
-
Spatial Memory by Agents, for Agents
arXiv cs.RO (Robotics)
arXiv:2610.22736v1 Announce Type: new Abstract: Spatial memory enables robots to reuse past observations for localization, navigation, and question answering. Designing such memory involves jointly deciding what to retain and how to…
-
Ego2Contact: Predicting Touch from Egocentric Human Video for Contact-Rich Robot Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.22932v1 Announce Type: new Abstract: Contact locations and interaction forces are critical for contact-rich robot manipulation but are often ambiguous from vision alone. Measuring them on robots requires specialized tactile…
-
Robotic Spatiotemporal Editing: A Unified Interface for Controllable Action Generation via Masked Action Reconstruction
arXiv cs.RO (Robotics)
arXiv:2610.23002v1 Announce Type: new Abstract: Generative policies can synthesize robot behaviors, yet controlling and editing them through diverse spatiotemporal references remains challenging. In the real world, human intent requires…
-
Unifying Embodied Foundation Models via a Model-to-Policy Formulation
arXiv cs.RO (Robotics)
arXiv:2610.23053v1 Announce Type: new Abstract: Vision-language-action (VLA) models and world-action models (WAMs) have emerged as promising approaches to robot learning and control. These models integrate multimodal perception and action…
-
Foundation-Model-Driven 3D Scene Graphs for Mobile Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.23622v1 Announce Type: new Abstract: Geometric and Semantic Foundation models are becoming increasingly popular tools for generating spatial maps for robots. Recent work (e.g., FOUND-IT) has leveraged these tools to construct…
-
Blind Precise Assembly: Proprioceptive Sim-to-Real Peg-in-Hole for Diverse Parts
arXiv cs.RO (Robotics)
arXiv:2610.23849v1 Announce Type: new Abstract: Every robot already carries high-resolution contact sensors: its own joints. Legged robots and humanoids exploit them to walk, run, and track whole-body motions without vision, but there the…
-
MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models
arXiv cs.RO (Robotics)
arXiv:2610.21686v2 Announce Type: replace Abstract: Reliable contact-rich manipulation requires detecting execution deviations early enough for corrective action. Yet most monitoring methods are reactive, identifying anomalies only after…
-
DefFlow: Goal-Conditioned Deformable Object Manipulation using 3D Point Flows
arXiv cs.RO (Robotics)
arXiv:2610.22827v1 Announce Type: new Abstract: Manipulating deformable objects toward desired 3D configurations is a fundamental capability for robots operating in unstructured physical environments. However, as the range of possible…
-
What Makes a Good Symbol for Hybrid Discrete-Continuous Planning?
arXiv cs.RO (Robotics)
arXiv:2610.23010v1 Announce Type: new Abstract: The physical world has continuous state and action spaces, yet robot planning often benefits from hybrid discrete-continuous formulations that explicitly expose combinatorial structure, such…
-
Bio-Inspired Magnetic Navigation Using Gradient Descent With Rao-Blackwellized Particle Filtering
arXiv cs.RO (Robotics)
arXiv:2610.23097v1 Announce Type: new Abstract: Navigation in GPS-denied environments is commonly based on magnetic maps, which are used to correct trajectory drift and enable relocalization. However, such navigation systems suffer from a…
-
OrbiSim: World Models as Differentiable Physics Engines for Embodied Intelligence
arXiv cs.RO (Robotics)
arXiv:2605.16395v2 Announce Type: replace Abstract: We present OrbiSim, a novel robotic simulation paradigm that redefines world models as a fully differentiable physics engine for embodied intelligence. Unlike prior world models that…
-
DexBioLab: Benchmarking Dexterous Manipulation with Dynamic Experimental Phenomena in Realistic Laboratories
arXiv cs.RO (Robotics)
arXiv:2610.13643v4 Announce Type: replace Abstract: Automating biological laboratories requires robots to perform dexterous tool interactions under realistic visual conditions while handling dynamic experimental phenomena, presenting…
-
Scaling Online Exploration for Robot Agents
arXiv cs.RO (Robotics)
arXiv:2610.22496v1 Announce Type: new Abstract: Robot learning has advanced by scaling demonstration datasets and model capacity; however, deploying a new robot embodiment or task typically necessitates a dedicated teleoperation corpus…
-
Structured Perception Representations for Off-road Dynamics Learning
arXiv cs.RO (Robotics)
arXiv:2610.22900v1 Announce Type: new Abstract: Autonomously navigating off-road terrain at high speed demands accurate control, which often relies on dynamics models that capture complex interactions between the robot and terrain. In…
-
GRASP: Global-Local Representation Alignment and Spatial-Physical Grounding for Bimanual Robotic Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.23123v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models that directly regress joint-space actions avoid the deployment-time cost of inverse kinematics. However, their training objectives remain misaligned with…
-
BeJEPA: A Goal-Conditioned Humanoid Behavior World Model via Unsupervised RL
arXiv cs.RO (Robotics)
arXiv:2610.23127v1 Announce Type: new Abstract: Humanoid control requires coordinating whole-body posture with position and heading in the world frame. Mainstream controllers imitate trajectories in the root frame, but struggle to regain…
-
LY-GWM-RoboCasa-Human300: Prediction-Guided Action Selection with a Graph Dynamics Model
arXiv cs.RO (Robotics)
arXiv:2610.23274v1 Announce Type: new Abstract: We present LY-GWM-RoboCasa-Human300, a policy that combines a Human300-fine-tuned GR00T-N1.5 action generator, an action-conditioned graph dynamics model, and a learned binary state-reward…
-
Enhancing Visual Task-Redirection Attacks on Vision-Language-Action Models via Closed-Loop Optimization
arXiv cs.RO (Robotics)
arXiv:2610.23566v1 Announce Type: new Abstract: Vision-language-action (VLA) models enable language-guided robot manipulation, but remain vulnerable to visual attacks that undermine user control over robot behavior. Prior studies optimize…
-
Self-Building Labs
arXiv cs.RO (Robotics)
arXiv:2610.23826v1 Announce Type: cross Abstract: Self-driving laboratories (SDLs) accelerate scientific discovery by automating experimental execution and optimization. AI agents can increasingly propose new experiments, but existing SDLs…
-
Reason-guided Human Reward Modeling for Preference-based Reinforcement Learning
arXiv cs.RO (Robotics)
arXiv:2610.23550v1 Announce Type: new Abstract: Reinforcement learning (RL) policies optimized for task rewards may produce behaviors that are efficient but misaligned with human expectations, safety preferences, or intent. This paper…
-
SparseMax-dSTL: Sound Robustness with Horizon-Independent Gradient Allocation for Temporal Logic Trajectory Optimization
arXiv cs.RO (Robotics)
arXiv:2610.23837v1 Announce Type: cross Abstract: A gradient-based solver that enforces a Signal Temporal Logic (STL) specification uses a smooth approximation of its quantitative semantics, the robustness. In safety-critical trajectory…
-
WorldOracle: Benchmarking Action-Conditioned World Models against Real-World Executions
arXiv cs.RO (Robotics)
arXiv:2610.18826v2 Announce Type: replace Abstract: We introduce WorldOracle, a benchmark specifically designed to evaluate action-conditioned world models (ACWMs) against paired real-world executions. Given the same initial observation…
-
MOSS: Multimodal Origami Swamp Robot System
arXiv cs.RO (Robotics)
arXiv:2610.23813v1 Announce Type: new Abstract: Robotic systems for environmental monitoring and ecological surveying in wetlands and swamps must operate across diverse terrestrial and aquatic conditions. Conventional fixed-geometry wheels…
-
OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism
arXiv cs.RO (Robotics)
arXiv:2603.14371v5 Announce Type: replace Abstract: Embodied AI agents increasingly require parallel execution of multiple tasks, such as manipulation, conversation, and memory construction, from shared observations under distinct time…
-
Cyclops: LiDAR as a Camera That Dreams in Color
arXiv cs.RO (Robotics)
arXiv:2608.16264v2 Announce Type: replace Abstract: Conventionally, robotic perception relies heavily on cameras due to the rich semantic texture they provide. However, their performance degrades significantly in low-light or…
-
HexaShell: A Thin Harness for Exploring the Limits of VLM Agents as Robot Coding Agents
arXiv cs.RO (Robotics)
arXiv:2610.23479v1 Announce Type: new Abstract: Agents built on vision language models (VLMs) now compete with learned policies on robot manipulation leaderboards. We ask how far a coding agent can go when it controls a robot instead of a…
-
ADEPT: Accelerating Dexterity via Pre-Training and Post-Training using Reinforcement Learning
arXiv cs.RO (Robotics)
arXiv:2608.19182v2 Announce Type: replace Abstract: We introduce Accelerating Dexterity via Pre-Training (ADEPT), a large-scale reinforcement learning (RL) framework for learning sim-to-real transferable dexterity across high…
-
WireCraft: A Simulation Benchmark for Industrial DLO Manipulation
arXiv cs.RO (Robotics)
arXiv:2606.18097v3 Announce Type: replace Abstract: Deformable Linear Objects (DLOs), such as wires and cables, are central to industrial assembly. Unlike rigid objects, whose state is captured by a 6-DoF pose, DLOs have an…
-
MambaAct: Efficient Mamba Action Experts for Flow-Matching Vision-Language-Action Models
arXiv cs.RO (Robotics)
arXiv:2610.22951v1 Announce Type: new Abstract: Vision-language-action (VLA) models have shown strong promise for generalist robotic manipulation, but their inference cost remains a major bottleneck for real-time control. This limitation…
-
PanoAir: A Panoramic Visual-Inertial SLAM with Cross-Time Real-World UAV Dataset
arXiv cs.RO (Robotics)
arXiv:2604.00852v2 Announce Type: replace Abstract: Accurate pose estimation is fundamental for unmanned aerial vehicle (UAV) applications, where Visual-Inertial SLAM (VI-SLAM) provides a cost-effective solution for localization and…
-
FELT: Generating Tactile Signals from Vision for Visuo-Tactile Manipulation
arXiv cs.RO (Robotics)
arXiv:2607.20683v2 Announce Type: replace Abstract: Tactile sensing can improve performance in contact-rich robot manipulation. However, learning policies that use real tactile feedback typically requires demonstrations collected with…
-
PRISM: Polynomial Representations for Interaction-Structured Motor Control
arXiv cs.RO (Robotics)
arXiv:2607.23473v3 Announce Type: replace Abstract: Robot policies map observations to actions, typically using multilayer perceptrons (MLPs). Yet robot dynamics and control objectives involve multiplicative interactions among observable…
-
Diagnosing Compositional Generalization in Sequential Robot Tasks
arXiv cs.RO (Robotics)
arXiv:2607.29687v2 Announce Type: replace Abstract: Sequential robot manipulation requires policies to execute novel combinations of familiar instruction components. However, collecting demonstrations for all possible instruction tuples is…
-
Same Goal, Different Words: Spatial Instruction Sensitivity in World-Action Models
arXiv cs.RO (Robotics)
arXiv:2610.16242v3 Announce Type: replace Abstract: World-action models should preserve the intended manipulation across equivalent descriptions of a spatial goal. We evaluate this consistency in 864 rollouts from three models across four…
-
GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives
arXiv cs.RO (Robotics)
arXiv:2609.39601v2 Announce Type: replace-cross Abstract: Precise grounding matters. It specifies which object is the target and where that object is, even in clutter and for tiny objects, and it has to be fast enough for closed-loop…
-
Towards Unified World Models for Visual Navigation via Memory-Augmented Planning and Foresight
arXiv cs.RO (Robotics)
arXiv:2510.08713v4 Announce Type: replace-cross Abstract: Enabling embodied agents to imagine future states is essential for robust and generalizable visual navigation. Yet, state-of-the-art systems typically rely on modular designs that…
-
All Roads Lead to Rome: Flow-driven Multi-Anchor Exploration for Open-Environment Active 3D Mapping
arXiv cs.RO (Robotics)
arXiv:2609.36889v2 Announce Type: replace Abstract: To advance the development of embodied intelligence, Open-Environment Active 3D Mapping has attracted increasing attention, aiming to perform a long-horizon and shortest trajectory…
-
When Plans Change, Remember: Replanning-Guided Visual Memory for Vision-Language-Action Models
arXiv cs.RO (Robotics)
arXiv:2610.22556v1 Announce Type: new Abstract: Long-horizon, multi-stage Vision-Language-Action (VLA) policies often require visual information that is no longer present in the current observation. Retaining the complete observation…
-
InstAgent: A Recursively Self-Improving Agent Operating System for Heterogeneous Robot Collaboration
arXiv cs.RO (Robotics)
arXiv:2610.23098v1 Announce Type: new Abstract: Heterogeneous robot teams enable complex, long-horizon tasks by combining complementary sensing, mobility, and manipulation capabilities across multiple embodiments. Yet existing systems…
-
Reinforcement Learning Based Multi-Terrain Adaptive Whole-Body Control for a Quadruped Manipulator
arXiv cs.RO (Robotics)
arXiv:2610.23548v1 Announce Type: new Abstract: Legged robots equipped with manipulators offer promising capabilities for loco-manipulation tasks in unstructured environments. However, achieving task-consistent end-effector behavior during…
-
TransNav: Transformer-Based End-to-End Navigation Method for Static and Dynamic Quadrotor Obstacle Avoidance
arXiv cs.RO (Robotics)
arXiv:2610.23635v1 Announce Type: new Abstract: Obstacle avoidance for autonomous quadrotors in cluttered environments remains challenging, especially when both dense static obstacles and moving obstacles are present. Existing methods are…
-
MANGO-Grasp: Mahalanobis Fields over Geometry-Oriented 3D Gaussians for Cross-Embodiment Dexterous Grasping
arXiv cs.RO (Robotics)
arXiv:2608.02014v2 Announce Type: replace Abstract: Cross-embodiment dexterous grasping aims to synthesize stable grasps across heterogeneous multi-fingered hands with little or no embodiment-specific tuning. Existing interaction-centric…
-
Can a Frontier VLM Serve as a Robot Policy for Memory-Dependent Manipulation?
arXiv cs.RO (Robotics)
arXiv:2610.17350v2 Announce Type: replace Abstract: We use a frozen frontier vision-language model, GPT-6-Astra, directly as the robot policy on RoboMME, a benchmark of 16 memory-dependent manipulation tasks spanning temporal, spatial…
-
Perceptive Behavior Foundation Model: Adapting Human Motion Priors to Robot-Centric Terrain
arXiv cs.RO (Robotics)
arXiv:2606.08059v3 Announce Type: replace Abstract: Humanoid behavior foundation models aim to acquire reusable whole-body control policies from broad human motion priors, enabling a single controller to produce diverse and expressive…
-
Analytic Robot Self-Model: Learning Screw-Axis Kinematics for Rendering and Model-Based Control
arXiv cs.RO (Robotics)
arXiv:2610.15558v2 Announce Type: replace Abstract: Visual self-modelling lets robots learn body models from observations, but such models usually encode kinematics implicitly rather than exposing metric joint parameters for computation…
-
Beyond Action Prediction: Learning Search-Free Planning in Latent Space
arXiv cs.RO (Robotics)
arXiv:2610.23615v1 Announce Type: new Abstract: Latent world models learn action-conditioned state transitions in a compact representation space, providing a predictive model for visual planning and control. However, planning with these…
-
Not All Actions Contribute Equally: Learning Precise Progress Estimation from Imprecise Demonstrations
arXiv cs.RO (Robotics)
arXiv:2610.23769v1 Announce Type: new Abstract: Progress estimators play an essential role in providing dense reward signals for reinforcement learning in robotic manipulation, but accurately assessing task progress can be much…
-
GR-VIP: Geometric representation conditioning for precise visuomotor policy
arXiv cs.RO (Robotics)
arXiv:2610.22997v1 Announce Type: new Abstract: Achieving precise terminal position and heading alignment remains a significant challenge for learning-based mobile robot control. Recent vision-based learning methods have shown promise for…
-
Expand-and-Prune: Test-Time Candidate-Space Adaptation for End-to-End Autonomous Driving
arXiv cs.RO (Robotics)
arXiv:2610.23262v1 Announce Type: new Abstract: Candidate-based end-to-end driving planners reason over a finite set of trajectory hypotheses. However, for sparse planners, a limited candidate set may provide insufficient support for…
-
EmbodiedMidtrain: Bridging the Gap between Vision-Language Models and Vision-Language-Action Models via Mid-training
arXiv cs.RO (Robotics)
arXiv:2604.20012v2 Announce Type: cross Abstract: Vision-Language-Action Models (VLAs) inherit their visual and linguistic capabilities from Vision-Language Models (VLMs), yet most VLAs are built from off-the-shelf VLMs that are not…
-
HydroShear: Hydroelastic Shear Simulation for Tactile Sim-to-Real Reinforcement Learning
arXiv cs.RO (Robotics)
arXiv:2603.00446v2 Announce Type: replace Abstract: In this paper, we address the problem of tactile sim-to-real policy transfer for contact-rich tasks. Existing methods primarily focus on vision-based sensors and emphasize image rendering…
-
When Does Corruption Break the Plan? A Degradation-Tolerance Benchmark for Camera-Only End-to-End Driving
arXiv cs.RO (Robotics)
arXiv:2608.29005v3 Announce Type: replace Abstract: Camera-only end-to-end (E2E) driving models are nearing deployment, where the camera stream is degraded by blur, noise, low light, weather, frame loss, and memory faults. How much a…
-
SkyAgent: Revisable Target Memory for Continuous Aerial Search
arXiv cs.RO (Robotics)
arXiv:2610.23293v1 Announce Type: new Abstract: Continuous aerial search builds target hypotheses from sparse, viewpoint-dependent observations. When later evidence contradicts a stored hypothesis, however, the candidate can remain…
-
Learning What to Say to Your VLA: Mostly Harmless Vision Language Action Model Steering
arXiv cs.RO (Robotics)
arXiv:2606.12299v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle and unintuitive: semantically similar…
-
Task Structure and Scalability in Multi-Robot Authoring
arXiv cs.RO (Robotics)
arXiv:2610.22670v1 Announce Type: new Abstract: Multi-robot authoring can reduce completion time by distributing drawing work, but the resulting speedup depends on task structure and the allocation-routing pipeline. We investigate these…
-
FLoTS-SLAM: Optical Flow-Derived Uncertainty for Monocular Dynamic SLAM with Two-Stage Optimization
arXiv cs.RO (Robotics)
arXiv:2610.23496v1 Announce Type: new Abstract: Monocular simultaneous localization and mapping (SLAM) remains challenging in dynamic environments. Real-time dynamic object identification in such environments is a critical problem. Recent…
-
NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Models
arXiv cs.RO (Robotics)
arXiv:2610.23271v1 Announce Type: new Abstract: Vision--Language--Action (VLA) models promise generalist robotic agents, yet consolidating skills learned across heterogeneous tasks and environments remains difficult: training data is…
-
CoEvolve: Self-Evolving Code Policies from Skills to Strategies for Humanoid Robots
arXiv cs.RO (Robotics)
arXiv:2610.23288v1 Announce Type: new Abstract: Recent advances in Vision-Language-Action (VLA) models and World Action Models (WAMs) have expanded the manipulation and whole body capabilities of individual robots. Yet a single robot…
-
Online Constraint Optimization-Based Indirect-Adaptive Robust Control for Autonomous Vehicles with Unknown Friction
arXiv cs.RO (Robotics)
arXiv:2610.22675v1 Announce Type: cross Abstract: Friction uncertainty is a significant challenge for autonomous vehicle control, undermining feasibility and performance. This paper proposes a novel two-layer control framework to achieve…
-
Mind the Phase: Effective Rank and Representation Health in Legged Locomotion
arXiv cs.RO (Robotics)
arXiv:2609.06958v2 Announce Type: replace Abstract: Reinforcement learning has become the leading paradigm in legged locomotion, enabling complex behaviors from backflips to parkour via massively parallel simulation. Under PPO's…
-
Embodiment-Classifier Guidance: Transferring Frozen VLAs Across Robot Dynamics without Demonstrations
arXiv cs.RO (Robotics)
arXiv:2610.22576v1 Announce Type: new Abstract: Flow-based policies such as $\pi_0$ are trained on a fixed set of embodiments, and transferring them to new ones is an open problem. To adapt a frozen $\pi_0$ to a new embodiment without…
-
RTISSLAM-3D: Full 6-DoF SLAM With an In-Air 3D Imaging Sonar
arXiv cs.RO (Robotics)
arXiv:2610.22552v1 Announce Type: new Abstract: Echolocating bats navigate dense forests using sonar alone, and in-air 3D imaging sonars now capture the entire frontal hemisphere with every pulse. Sonar SLAM, however, has so far been…
-
AWAM: Active Trajectory-Conditioned View Imagination for Robust Robot Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.23396v1 Announce Type: new Abstract: Rapid progress in embodied intelligence has enabled increasingly capable robots to follow language instructions and manipulate diverse objects. However, strong results in structured…
-
VLA-Fabric: Learning Interaction for Composable Multi-Arm VLA Agents
arXiv cs.RO (Robotics)
arXiv:2610.23018v1 Announce Type: new Abstract: Vision-language-action (VLA) models are rapidly expanding what individual robots can accomplish, yet many real-world manipulation settings require multiple arms to operate as a coordinated…
-
Foundation Models for Autonomous Space Robotics: Architectures, Physical Adaptation, and Validation
arXiv cs.RO (Robotics)
arXiv:2610.23068v1 Announce Type: new Abstract: Space robots must interpret tasks and act despite unfamiliar dynamics, difficult sensing conditions, and limited opportunities for human intervention. Foundation models offer reusable…
-
Sample-Efficient Reinforcement Learning for Policy Improvement via Flow-Reversed Action Anchors
arXiv cs.RO (Robotics)
arXiv:2610.22852v1 Announce Type: new Abstract: Generalist robot policies offer strong behavioral priors, but achieving deployment-level reliability and throughput remains difficult. Reinforcement learning (RL) can improve these policies…
-
Structural Affinity: An Algorithm for Coherent Cage-Based Shape Design
arXiv cs.RO (Robotics)
arXiv:2610.23173v1 Announce Type: cross Abstract: Cage-based deformation (CBD) parametrises 3D geometry through a small set of control handles, effectively enabling a variety of tasks from animation and interactive modelling to tool-based…
-
Beyond Success Rate: Interventional Evaluation of Memory in Vision-Language-Action Systems
arXiv cs.RO (Robotics)
arXiv:2610.23372v1 Announce Type: new Abstract: When we say that a robot uses memory, does that memory actually play a role? Success-rate ablations alone cannot establish the use of task-relevant facts: they change information and…
-
Learning Robot Metareasoning: Adaptive Computation Budgets for Real-Time Control
arXiv cs.RO (Robotics)
arXiv:2610.23662v1 Announce Type: new Abstract: Planning while moving faces an historical trade-off between deliberation and reactivity. A planner running concurrently with execution needs time to compute, so the robot must commit to an…
-
Shape-Aware Neural Time Fields for Rigid-Body Motion and Manipulation Planning
arXiv cs.RO (Robotics)
arXiv:2610.22762v1 Announce Type: new Abstract: Rigid-body motion planning is fundamental to robotic transport and manipulation, where both object position and orientation determine whether a body can move through cluttered environments…
-
SimpleVLN: Towards Simple and Frontier Vision-Language Navigation
arXiv cs.RO (Robotics)
arXiv:2610.23203v1 Announce Type: new Abstract: Vision-and-Language Navigation (VLN) requires embodied agents to ground language in visual observations and make long-horizon decisions. Existing methods often address these demands through…
-
Multi-Agent Reinforcement Learning-Based UAV Pathfinding for Obstacle Avoidance in Stochastic Environment
arXiv cs.RO (Robotics)
arXiv:2310.16659v3 Announce Type: replace Abstract: Traditional methods plan feasible paths for multiple agents in the stochastic environment. However, the methods' iterations with the changes in the environment result in computation…
-
Simultaneous Force and Motion Control in Generalised Coordinates
arXiv cs.RO (Robotics)
arXiv:2610.22813v1 Announce Type: new Abstract: Robot manipulators often come into contact with rigid objects when performing work. This imposes physical constraints on their motion. Simultaneous force & motion control was developed to…
-
ORD-WM: Learning Object-Centric Rigid Dynamics for 3D World Modeling in Robotic Manipulation
arXiv cs.RO (Robotics)
arXiv:2610.23156v1 Announce Type: new Abstract: World models provide robots with a predictive mechanism for reasoning about how actions change object states in manipulation tasks. However, existing image-based and dense 3D world models do…
-
NativeMEM: Native Memory Compression for Long-Horizon Robotic Manipulation
arXiv cs.RO (Robotics)
arXiv:2607.06678v2 Announce Type: replace Abstract: How can pretrained Vision-Language-Action (VLA) models retain long-horizon visual histories with high-frequency updates without sacrificing efficiency? Existing approaches rely on…
-
Hardware-as-Regularization: Enhancing Data Consistency in Imitation Learning via an Adjustable Passive Compliant Leader Arm
arXiv cs.RO (Robotics)
arXiv:2610.22977v1 Announce Type: new Abstract: Low-cost passive leader arms enable scalable data collection for imitation learning, but their lack of restoring forces leaves redundant joints unconstrained, yielding highly inconsistent…
-
SenseSwarm: Swarm Exploration via Neural Structure Estimation with Shared World-Anticipation for Rapid Mapping
arXiv cs.RO (Robotics)
arXiv:2610.22905v1 Announce Type: new Abstract: A robot swarm covers a building faster with more robots, but the map is only as complete as the space the robots have physically visited. We target the amount of an unknown building a…
-
No Eyes, No Problem: What Does Touch Contribute to Blind Baoding Ball Manipulation?
arXiv cs.RO (Robotics)
arXiv:2610.23438v1 Announce Type: new Abstract: Humans manipulate objects within their hands through touch alone with remarkable ease, but for robots, achieving sustained real-world manipulation without vision remains an open challenge…
-
EgoTac: In-the-wild Tactile Prediction from Egocentric Vision
arXiv cs.RO (Robotics)
arXiv:2608.15060v2 Announce Type: replace-cross Abstract: Touch is fundamental to dexterous manipulation, yet most egocentric human data increasingly used for robot learning lacks tactile information. Directly collecting large-scale…
-
Enabling Tidying Tasks in Robotic Vacuum Cleaners via a Geometry-Induced Self-Locking Deployable Linkage
arXiv cs.RO (Robotics)
arXiv:2610.20150v2 Announce Type: replace Abstract: Robotic vacuum cleaners must extend their functionality within tight limits on size, cost, and durability. Deployable linkages with few actuators meet these limits, but their…
-
Schr\"odinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation
arXiv cs.RO (Robotics)
arXiv:2512.21201v5 Announce Type: replace Abstract: Zero-shot object navigation (ZSON) requires robots to find target objects in unseen environments without task-specific fine-tuning or pre-built maps, a key capability for general-purpose…
-
FCTS: Failure-Guided Curriculum Tree Search Over Generated Terrain for Legged Parkour
arXiv cs.RO (Robotics)
arXiv:2610.22619v1 Announce Type: new Abstract: Quadruped parkour policies handle canonical obstacles well, but their training curricula often miss the irregular, compound geometry of industrial and disaster-response sites, where agile…
-
Benchmarking Robots for Everyday Environments: From Lab Experiments to Real-World Operations
arXiv cs.RO (Robotics)
arXiv:2609.26490v2 Announce Type: replace Abstract: This study presents an interdisciplinary process for benchmarking service robots in public everyday environments, addressing the gap between laboratory metrics and real-world operation at…
-
Monkey See, Can Monkey Do? A Benchmark for Evaluating Robot Skill Learning by Observation
arXiv cs.RO (Robotics)
arXiv:2609.08209v4 Announce Type: replace Abstract: Learning from Observation (LfO) is a fundamental robotic capability that replicates how humans and animals socially learn from each other. Beyond its biological parallels, this modality…
-
HyVGGT-VO: Hybrid Dense Visual Odometry With Classical VO and Feed-Forward Model
arXiv cs.RO (Robotics)
arXiv:2604.02107v2 Announce Type: replace Abstract: Dense visual odometry (VO), which provides pose estimation and dense 3D reconstruction, serves as the cornerstone for applications ranging from robotics to augmented reality. Recently…
-
SPACE: Swarm Pheromone Fields for Adaptive Collision-Aware Exploration
arXiv cs.RO (Robotics)
arXiv:2606.29372v2 Announce Type: replace Abstract: Massive robot swarms can explore unknown environments quickly, but adding robots eventually stops helping. Doorways and dense traffic create congestion, increasing inter-robot contacts…
-
STRIPS-WM: Learning Grounded Propositional STRIPS-style World Models from Images
arXiv cs.RO (Robotics)
arXiv:2606.06832v2 Announce Type: replace Abstract: Robots performing long-horizon visual manipulation observe high-dimensional images, but successful plans depend on action-relevant facts: what can be done now and what changes afterward…
-
Micro-Localization: High-Precision Mobile Robot Localization Using Microscopic Ground Texture
arXiv cs.RO (Robotics)
arXiv:2610.22563v1 Announce Type: new Abstract: We demonstrate high-precision global localization using natural ground texture viewed through microscope optics, achieving localization in the tens-of-micrometers regime. For the planar…
-
CorrelationFlow: A Training-Free Geometric Approach for LiDAR Scene Flow Estimation
arXiv cs.RO (Robotics)
arXiv:2607.29237v3 Announce Type: replace-cross Abstract: LiDAR scene flow is the task of estimating point-wise motion from sequential point clouds. It is essential for autonomous driving, as it provides critical information about moving…
-
LAW: Unifying Language, Action, and World Modeling for Agentic Physical Intelligence
arXiv cs.RO (Robotics)
arXiv:2610.23334v1 Announce Type: new Abstract: World-action models jointly predict actions and future world states, but long-horizon manipulation further requires language reasoning to plan subtask objectives, assess execution progress…
-
NEXUS: Learning Safety Constraints and Action Models for Safe and Robust Embodied Planning
arXiv cs.RO (Robotics)
arXiv:2605.09387v2 Announce Type: replace-cross Abstract: Large language models (LLMs) enable embodied agents to plan from naturallanguage instructions, yet their actions may be physically infeasible or hazardous. Existing neuro-symbolic…
-
DriveWorld3D Evidence Constrained 3D World Modeling and Generative Simulation for Real-World Autonomous Driving
arXiv cs.RO (Robotics)
arXiv:2610.22450v1 Announce Type: new Abstract: Three dimensional modeling for autonomous driving must continuously estimate road geometry, obstacle occupancy, traffic participant behavior, and observational uncertainty under limited…
-
From Sliding Windows to Self-evolving Lifelong Memory: A Dual-perspective Survey of Memory Paradigms for Long-Horizon Robotics VLA Models
arXiv cs.RO (Robotics)
arXiv:2610.23319v1 Announce Type: new Abstract: Long-horizon robotic tasks expose a fundamental limitation of current Vision-Language-Action (VLA) models because their restricted temporal context pre-vents effective accumulation and reuse…
-
PASS-VLN: Passability-Aware Local-Command Vision-and-Language Navigation
arXiv cs.RO (Robotics)
arXiv:2610.22812v1 Announce Type: new Abstract: Reliable vision-and-language navigation (VLN) requires not only understanding instructions but also determining whether and how a robot can physically execute them. However, existing VLN…
-
PASS: A Trustworthy Precision Approach Simulation System for CAT III Integrity Algorithm Development
arXiv cs.RO (Robotics)
arXiv:2610.23019v1 Announce Type: new Abstract: For fixed-wing aircraft, reliable localization during approach and landing is critical to flight safety. Visual--inertial navigation is promising, but aircraft navigation also requires…
-
Look Before You Act: Bridging Active Perception and Vision-Language-Action Models via Observation-Aligned Data Adaptation
arXiv cs.RO (Robotics)
arXiv:2610.23378v1 Announce Type: new Abstract: Robots manipulating objects in clutter must decide not only how to act, but also where to look. Although active camera motion can expose an occluded target, it may move the robot into an…
-
F-LIVO: A Feature-Level Geometric Measurement Framework for LiDAR-Inertial-Visual Odometry
arXiv cs.RO (Robotics)
arXiv:2610.23595v1 Announce Type: new Abstract: Measurement-level LiDAR-Inertial-Visual Odometry (LIVO) can achieve high accuracy, and its practical resource requirements depend on the adopted association and map-maintenance strategy…