MEKAWIRE
Research

Papers

New robotics and physical-AI papers, with what each one means for the industry.

276 papers
TopicAllManipulation · 27VLA · 22Perception · 12Robotics · 11Navigation · 10Humanoid · 9Learning · 8Safety · 5World Action Models · 4Simulation · 3Aerial manipulation · 2Dexterous manipulation · 2Sim-to-Real Manipulation · 2Social Navigation · 2Vision-Language Navigation · 2World models · 2World-action model · 23D Assembly · 13DGS dataset · 15G Simulation · 1AMR Operations · 1Active vision manipulation · 1Adaptive MPC · 1Adaptive action horizons · 1Adaptive intervention · 1Aerial · 1Aerial Manipulation · 1Aerial navigation · 1Agricultural Robotics · 1Articulation Estimation · 1Autonomous Driving · 1Autonomous Exploration · 1Battery-aware RL · 1Belief-space planning · 1Bimanual Manipulation · 1Byzantine localization · 1Canal inspection · 1Co-speech Humanoid · 1Compositional Manipulation · 1Construction navigation · 1Contact-Rich Manipulation · 1Contact-rich Manipulation · 1Continuum sensing · 1Curriculum RL · 1Dexterous Manipulation · 1Distributed Manipulation · 1Distribution Shift · 1Driving Simulation · 1Efficient VIO · 1Egocentric data · 1Egocentric video · 1Embodied Agents · 1Embodied policy · 1Empowerment MPC · 1Endoscopic navigation · 1Endovascular robotics · 1Event-camera VO · 1Exoskeletons · 1Failure recovery · 1Federated 3D Perception · 1Fisheye detection · 1Floorplan SLAM · 1Geometry-aware manipulation · 1Grasping · 1Grounding · 1Humanoid Control · 1Humanoid RL · 1Humanoid Tool Use · 1Humanoid control · 1Humanoid loco-manipulation · 1Humanoid safety · 1Imitation Learning · 1JEPA World Models · 1Lab automation · 1Laboratory Manipulation · 1Latent World Models · 1Legged Manipulation · 1LiDAR Security · 1Lifelong RL · 1Locomotion · 1MAPF · 1MAV guidance · 1Manipulation LfD · 1Manipulation Safety · 1Marine simulation · 1Memory-centric VLA · 1Miniature Locomotion · 1Mining Perception · 1Mobile manipulation · 1Model-based RL · 1Monocular SSC · 1Motion Planning · 1Multi-Agent RL · 1Multi-Robot Exploration · 1Multi-Robot Manipulation · 1Multi-UAV Exploration · 1Multi-UAV State Estimation · 1Multi-agent RL · 1Multi-agent motion planning · 1Multi-camera odometry · 1Multi-fidelity RL · 1Multi-robot manipulation · 1Multi-skill Manipulation · 1Multi-view perception · 1Multimodal simulation · 1Navigation & Mapping · 1Object Pose Estimation · 1Object Search · 1Panoramic 3D Detection · 1Panoramic segmentation · 1Pedestrian tracking · 1PersonNav · 1Planning · 1Policy · 1Policy Optimization · 1Preference Learning · 1Quadruped locomotion · 1Real-time VLA · 1Real-to-sim · 1Representation alignment · 1Robot-Use Agents · 1Runtime Verification · 1Safe Generative Planning · 1Safe RL-MPC · 1Safe VLA · 1Safety monitoring · 1Safety-critical control · 1Sim-to-real Navigation · 1Sim-to-real manipulation · 1Simulation geometry · 1Skill Discovery · 1Social navigation · 1Soft Pneumatic Actuators · 1Soft manipulation · 1Sonar SLAM · 1Streaming 3D reconstruction · 1Submovement decomposition · 1Surgical robotics · 1Tactile · 1Tactile Exploration · 1Tactile VLA · 1Temporal manipulation · 1Terrain traversability · 1Text-to-3D Policy · 1Tool Co-Design · 1Trajectory Correction · 1Traversability · 1UAV Navigation · 1UAV Swarms · 1UAV traffic monitoring · 1Underwater Robotics · 1Underwater active mapping · 1Underwater perception · 1Unified world-action model · 1VLA Conditioning · 1VLA Defense · 1VLA Robustness · 1VLA evaluation · 1VLA manipulation · 1VLA robustness · 1Video world models · 1Visual Generalization · 1Visual Grounding · 1Visual Reward Models · 1Visuomotor Fusion · 1Visuotactile manipulation · 1World Models · 1World-Action Models · 1
LevelAnyBeginnerIntermediateExpert
arXiv
ManipulationIntermediate

Bridging Frontier Reasoning and Robot Execution: From Autonomous Demonstration Generation to Dense Language Supervision

Bosung Kim, Alexander Trevithick, Ruiyi Wang, Prithviraj Ammanabrolu

Preprint. The authors present two ways to connect slow frontier reasoning models to fast local robot policies: autonomous generation of corrective demonstrations for training, and dense language supervision at multiple granularities. They report that autonomous demonstration generation becomes more reliable and cheaper as successful examples accumulate, and that a deployment harness lets the frontier model guide the local policy. This matters because it could enable powerful but high-latency models to drive real-time robot manipulation.

Read the full paper (PDF)
arXiv
Contact-rich ManipulationIntermediate

RATE: Risk-Aware Tactile Encoding for Contact-rich Robotic Manipulation

Yuyao Jiang, Haichao Liu, Jiarui Zheng, Zihan Ding, Weihao Yuan, Ziwei Wang

RATE (preprint) introduces a risk-aware tactile encoding method that learns task-conditioned contact risk from tactile history and alert supervision, then injects it into a frozen ACT policy through a lightweight residual adapter. It reports 71.8% average success on eight UniVTAC simulation tasks and 73.3% average success on three real-world tasks, outperforming baselines including 3.3B-5B parameter visuomotor policies while updating only about 0.59% of model parameters. This matters for contact-rich manipulation because similar tactile readings can hide very different failure risks.

Tactile Sensing
Read the full paper (PDF)
arXiv
Distribution ShiftExpert

Self-Repairing Recurrent Ensembles for Real-Time Recovery from Distribution Shift

Julian Lemmel, Pedro D. Wendel Garcia, Taisuke Kobayashi, Radu Grosu

This preprint presents a self-repairing recurrent ensemble policy that recovers from sensor shifts online without expert labels. The headline result is that random observation masking plus Kalman fusion and per-step RFLO updates lets simulated locomotion policies regain near pre-shift performance after a sudden sensor shift, while full-visibility ensembles collapse and do not recover. The work matters for unattended robots that must self-correct when sensors drift, fail, or become noisy.

Read the full paper (PDF)
arXiv
World Action ModelsExpert

XGenAct: Geometry-Enhanced World Action Models through Cross-Task Generation

Tingting Du, Ziyao Wang, Guoheng Sun, Ang Li

XGenAct is a preprint describing a world action model that turns RGB observations, robot actions, metric depth, surface normals, and functional-role segmentation into RGB video streams and trains one video diffusion transformer on all of them. On five held-out RLBench tasks it averages 52% closed-loop success versus 26% for the strongest evaluated baseline. This matters because it adds explicit geometric and semantic future prediction to robot action generation without modality-specific heads or auxiliary losses.

Read the full paper (PDF)
arXiv
VLAExpert

FastOPD: On-Policy Distillation for Lightweight VLA Deployment

Yoojin Oh, Jeongsol Kim, Yeonwoo Seo, Jangho Park, Seonghyun Jin, Sunwoo Park, Youngmin Kim, Youngjun Jun, Kyumin Choi, Jong Chul Ye

This preprint introduces FastOPD, a method for distilling large vision-language-action (VLA) and world-action-model teachers into compact few-step robot policies. It keeps 84% of a 3.3B-parameter teacher's LIBERO success rate with only two sampling steps, reduces inference latency by 78.1%, and on RoboTwin 2.0 lifts single-step SmolVLA success from 35.3% to 51.2%. The work matters because it may allow foundation-policy capabilities to run on resource-constrained, low-latency robot hardware.

VLA
Foundation
Read the full paper (PDF)
arXiv
ManipulationIntermediate

AdaTempo: Learning Shared Relative Tempo from Demonstrations for Faster Robot Manipulation

Jiale Cao, Yike Niu, Zhengrong Xue, Huazhe Xu

This preprint presents AdaTempo, a self-supervised method that speeds up visuomotor imitation policies by extracting shared phase-relative tempo patterns from demonstrations and resampling trajectories before training. It achieves up to a 3.57x speedup over demonstrations, and in real-robot tests raises aggregate success from 58.3% with ACT to 77.8% while completing tasks 2.02x faster. The approach matters because it embeds faster execution directly into standard policies such as ACT or Diffusion Policy without runtime retiming or extra speed-selection modules.

Imitation Learning
Read the full paper (PDF)
arXiv
Safety monitoringExpert

Localized Conformal Safety Monitoring with Vision-Language Models for Autonomous Driving

Luís Marques, Rong Fang, Disha Kamale, Dmitry Berenson

This preprint introduces SLLCP, a post-hoc calibration layer over frozen vision-language models that turns raw safe/unsafe guesses into probabilistic safety prediction sets for autonomous driving. On 15,321 unseen CARLA frames, SLLCP flags 89.6% of collision-causing trajectories with Qwen3-VL-8B and 88.4% with Cosmos-Reason2-8B, versus 4.6% and 39.1% for the raw binary VLMs. It matters because it offers a planner-agnostic way to reduce missed unsafe trajectories before execution, albeit with more false alarms.

Read the full paper (PDF)
arXiv
VLA evaluationIntermediate

ManiPhysicsBench: Physics-Based Assessment of Object Preservation in VLA Manipulation

Sangwu Park, Yeonjun In, Wonjoong Kim, Sungwon Kim, Sein Kim, Chanyoung Park

This preprint introduces ManiPhysicsBench, a simulation-based benchmark that scores vision-language-action manipulation policies on object preservation, not just task completion. It combines a curated 91-object material property library with post-rollout physics-solver damage checks in LIBERO and SimplerEnv; public VLAs average 13.5% safe success versus 48.3% task success on one platform. The work matters because real-world deployment of fragile-object handling depends on not damaging goods, yet current benchmarks ignore this risk.

VLA
Read the full paper (PDF)
arXiv
Representation alignmentExpert

On Representational Alignment among Embodied Agents

Fulvio Mastrogiovanni

This preprint formalizes a task-dependent stopping condition for aligning representations across embodied agents. In an analytical hand-off scenario, the same ambiguity needs zero added sensorimotor anchors for relative synchronization but exactly one for absolute timing, because only the interaction invariance changes. This matters for multi-robot and human-robot systems: it suggests targeting alignment effort only where unresolved differences can change the task outcome.

Read the full paper (PDF)
arXiv
Monocular SSCExpert

CrowdOcc: Monocular Semantic Scene Completion for Quadruped Robots in Crowded Indoor Environments

Feiyang Chen, Jincheng Hu, Yiduo Chen, Jihao Li, Yue Liang, Bingzhao Gao, Yanjun Huang, Yuanjian Zhang

The authors present CrowdOcc, a preprint introducing an RGB-D dataset of 25.1K quadruped-captured frames in crowded indoor scenes and a monocular semantic scene completion model with surface-normal fusion and sparse human-centric 3D interaction. On held-out scenes it reaches 15.80 IoU, 11.40 mIoU, and 46.23 Human IoU, outperforming prior baselines and improving human, wall, and floor reconstruction. This matters because onboard single-camera dense 3D occupancy for low-viewpoint robots in cluttered human environments remains underexplored.

Read the full paper (PDF)
arXiv
Byzantine localizationExpert

Who Went Where When on the Lunar Surface: Forensic Trajectory Analysis to Identify Byzantine Rovers

Lachlan Holden, Feras Dayoub, Melissa de Zwart, David Harvey, Tat-Jun Chin

This preprint introduces a forensic trajectory analysis method for co-located lunar rover missions where some rovers may provide miscalibrated or deliberately falsified telemetry. An attribution-aware credibility method rejects whole non-credible rovers rather than individual outlier measurements and reconstructs trajectories from the remaining sources. In simulation it recalled up to 90.8% of non-credible measurements versus under 8% for robust pose-graph baselines, and on planetary-analogue data it was the only method to unambiguously expose a safety-zone incursion.

Read the full paper (PDF)
arXiv
TraversabilityIntermediate

From Language Priors to Field Adaptation: Preference Learning for Traversability Estimation

Simon Schwaiger, David Seyser, Alessandro Scherl, Zlatan Ajanović, Wilfried Wöber, Gerald Steinbauer-Wagner

This preprint presents a traversability estimator that models terrain preferences as von Mises-Fisher prototypes in a frozen vision-language feature space, initialized from language rules and fine-tuned with sparse relative image annotations. It achieves competitive accuracy with end-to-end trained models on WayFAST, e.g., HDR 0.5 of 21.8% vs 22.5% for the SOTA, while improving sample efficiency and using only about 8.2k trainable parameters. This matters because it enables rapid domain adaptation for outdoor mobile robots without extensive per-platform annotation.

Read the full paper (PDF)
arXiv
PersonNavExpert

CSIR: Contextually and Socially Informed Robots for Efficient Person Goal Navigation

Tyler Chung, Nahl Farhan, Hao Zhang, Mingfeng Yuan, Jinjun Shan, Amy Wu, Yue Hu

This preprint introduces CSIR, a planning framework for indoor person-goal navigation that combines semantic likelihoods from uncertain natural language requests with distance-aware route optimization. In synthetic benchmarks across 1,500 scenarios in three environments, CSIR achieved an average travel efficiency of 0.62, above the best classical baseline at 0.43 and comparable LLM-TRP at 0.60. It matters because it reduces dependence on private activity logs and fixed schedules, making service robots more deployable under realistic privacy and information-uncertainty constraints.

Read the full paper (PDF)
arXiv
Runtime VerificationExpert

Mind the Refinement Gap: When Safe High-Level Robot Plans Produce Unsafe Executions

Stabak Das, Priyesh Ranjan, Xiangfang Li, Lijun Qian

This preprint audits a trace-completeness assumption in RoboGuard by comparing safety verdicts on surface plans and graph-refined traces under the same LTL specification. In 12 controlled targeted cases the surface plan was accepted while the refined trace was rejected, and in end-to-end SPINE-RoboGuard tests 2 false-safe cases occurred. The work motivates lightweight graph-based trace refinement as a fix for formal safety monitoring of language-enabled planners.

Read the full paper (PDF)
arXiv
World modelsIntermediate

DeltaWorld: Physically Consistent Interactive World Simulators via Action-Conditioned Latent Increment Learning

Boyuan Hou, Xiaoge Cao, Chaofan Zhang, Shuo Wang, Shaowei Cui

DeltaWorld is a preprint describing an interactive world simulator for robot manipulation that predicts action-conditioned latent increments instead of full future latent states. On a self-collected cross-robot dataset it reduces FVD by 46.6% and LPIPS by 31.1% relative to the IWS baseline, and on the IWS benchmark it improves FID from 44.16 to 36.04. This matters because more stable long-horizon action-conditioned video prediction could reduce real-robot rollouts for policy training and evaluation.

World Models
Read the full paper (PDF)
arXiv
Miniature LocomotionExpert

Around the World: Unified Learned Locomotion on a 270 g Continuous-Rotation Quadruped

Arturo Flores Alvarez, Nathan Lintu, Dennis Hong

This preprint presents a single learned locomotion policy for MiNI-Q, a 270 g quadruped with continuous-rotation legs, running entirely onboard at 50 Hz. It reports forward-velocity RMSE of 0.037 m/s upright and 0.050 m/s inverted, 24/30 release recoveries, and successful walking with four foot geometries across four surfaces without retraining. The result shows that continuous-rotation topology can unify upright/inverted locomotion, posture changes, and recovery in one compact controller for sub-kilogram robots.

Read the full paper (PDF)
arXiv
Adaptive MPCIntermediate

LLA-MPC on Embedded Hardware: Rapid Adaptive Control with Thousands of Parallel Models

Henry Z. Liao, Maitham F. AL-Sunni, John M. Dolan

This preprint describes a generalized, learning-free adaptive model-predictive controller that runs thousands of candidate physics models in parallel on embedded hardware. On an F1TENTH car with low-friction tires, it completes high-speed tracking and changing-surface laps where a fixed nominal model crashes within a few laps. The work matters because it shows rapid online system identification and control can run on a small robot computer without pretraining or warm-up data.

Read the full paper (PDF)
arXiv
Humanoid RLIntermediate

Beyond Reward Hacking: Proxy Divergence Across Four Layers of a Staged Humanoid Learning Pipeline

Arunabh Bora

This preprint introduces a four-layer 'proxy audit' for humanoid reinforcement-learning pipelines, treating reward terms, curriculum gates, evaluation metrics, and reference motions as failure-prone proxies. The authors demonstrate measured failures and corrections in a simulated 1.91 m humanoid trained with PPO across four stages and 13,500 iterations on one laptop GPU; among the results, an averaged-error curriculum gate advanced on schedule while the target skill was missing, and a phase-aliased push test rated 0.5 m/s as more dangerous than 2.0 m/s. The low-cost reformulations matter because they can make simulated robot training results more trustworthy without retraining.

Humanoid
Read the full paper (PDF)
arXiv
Visuomotor FusionIntermediate

Register-Routed Delayed Fusion: Rewiring Shortcut-Prone Observation Fusion in Visuomotor Imitation

Jieting Long, Weidong Cai, Weiming Zhi

This arXiv preprint introduces Register-Routed Delayed Fusion (RRDF), a Transformer observation-fusion schedule for visuomotor imitation that blocks direct proprioception-to-vision attention and routes cross-modal information through learned registers. On three real-robot tasks, RRDF improves nominal success rates from 55%, 65%, and 50% for dense ACT to 100%, 100%, and 85%; under held-out object positions, it improves from 30%, 30%, and 0% to 95%, 100%, and 20%. The policy also shows lower state-to-image action sensitivity, suggesting more robust visual responsiveness in imitation-learned manipulation.

Read the full paper (PDF)
arXiv
Panoramic segmentationExpert

EmbPASS: Towards Cross-Embodiment Open Panoramic Segmentation

Pujun Guo, Yuanfan Zheng, Fei Teng, Mengfei Duan, Guoqiang Zhao, Yuheng Zhang, Kai Luo, Kailun Yang

The authors introduce EmbPASS, a 1,000-image panoramic semantic segmentation benchmark spanning vehicle, drone, wearable and quadruped platforms, and EPONet, a CLIP-based open-vocabulary segmentation network for cross-embodiment panoramic perception. On EmbPASS, EPONet reaches 35.82% mIoU, 1.10 points above the strongest baseline, and also leads DensePASS. This is a preprint.

Read the full paper (PDF)

Topic pages

Manipulation · 27VLA · 21Perception · 12General robotics · 11Navigation · 10Humanoid · 9Robot learning · 8Safety · 5World Action Models · 5Simulation · 3Aerial manipulation · 3Dexterous manipulation · 3Sim-to-Real Manipulation · 3Social Navigation · 3

Papers come from arXiv robotics feeds. Where our summary is not written yet, you see the opening lines of the abstract.