Dokumenti ta’ Riċerka dwar ir-Robotika — Paġna 23

How Instruct-Particulate Enables Fast 3D Object Articulation from Vision-Language Models

We scale feed-forward 3D object articulation by using vision-language models to pseudo-label synthetic 3D assets, enabling a transformer-based architecture t…

Ruining Li, Yuxin Yao, Matt Zhou, Chuanxia Zheng +4

RoboticsĠun 14

Navigation World Action Model for Goal-Conditioned Visual Navigation

NavWAM learns a joint world-action prediction model that turns visual foresight into a closed-loop policy for goal-conditioned visual navigation.

Daichi Azuma, Taiki Miyanishi, Koya Sakamoto +6

RoboticsĠun 14

SPARC: Reliable Spatial Annotations from Robot Demonstrations at Scale

Given a robot demonstration video, a language instruction, and gripper proprioceptive signal, SPARC produces object-centric spatial annotations for each inte…

Nils Blank, Paul Mattes, Maximilian Xiling Li +5

AIĠun 14

Heterogeneous LiDAR Early Fusion and Learned Re-Ranking for Robust Long-Term Place Recognition

Judith Vilella-Cantos, Juan José Cabrera, Mónica Ballesta +2

AIĠun 13

MaskWAM Teaches Robots to Predict What Matters for Better Manipulation

Hanyang Yu, Haitao Lin, Jingbo Zhang +4

AIĠun 13

LabVLA Brings Vision-Language-Action Models to Lab Automation

LabVLA achieves high success rates in lab tasks using synthetic data from RoboGenesis and transfers to real robots, advancing lab automation.

Baochang Ren, Xinjie Liu, Xi Chen +15

RoboticsĠun 13

MCR-Bionic Hand: How Anatomical Structure Unlocks Dexterous Robot Manipulation

Researchers have built MCR-Bionic, a musculoskeletal robotic hand that mimics key anatomical structures from the human hand—bones, ligaments, tendons, and in…

Haosen Yang, Guowu Wei

RoboticsĠun 12

Weaver: A Fast and Scalable World Model for Robotic Manipulation

Arnav Kumar Jain, Yilin Wu, Jesse Farebrother +2

RoboticsĠun 12

Flow Reversal Steering Makes Robot Generalists Better at Following High-Level Guidance

Andy Tang, William Chen, Andrew Wagenmaker +2

AIĠun 12

Mana: A New System for Dexterous Robot Manipulation of Articulated Tools

Mana trains robot hands to manipulate articulated tools like syringes and scissors using only simulation data, achieving 86.7% real-world success with zero sim-to-real retraining.

Zhao-Heng Yin, Guanya Shi, Pieter Abbeel +1

AIĠun 11

When and Where Should You Allocate Test-Time Compute in Embodied Planners?

Scaling test-time compute is not free in robotics: in a hierarchical stack the planner is invoked repeatedly, so per-call costs from chain-of-thought reasoni…

Jadelynn Dao, Milan Ganai, Yasmina Abukhadra +7

RoboticsĠun 11

World Pilot Gives Vision-Language-Action Robots a World-Action Model Boost

Zefu Lin, Rongxu Cui, Junjia Xu +4

AIĠun 11

New Method Gives Cheap Robot Arms Force Feedback Without Extra Sensors

Steven Oh, Jason Jingzhou Liu, Tony Tao +5

HumanoidĠun 10

Embodied mmWave Radar on Humanoid Robots Enables 24/7 Fall Detection

Researchers have developed EM-Fall, a system that mounts mmWave radar on a humanoid robot to detect falls day and night, even in complete darkness. The robot…

Yanshuo Lu, Yuxuan Hu, Shenghai Yuan +5

RoboticsĠun 10

Bimanual Joining System Lets Wheelchair Robots Call for Backup

Researchers at Northeastern University developed JOIN, a three-phase system that enables a mobile robot to autonomously join a wheelchair-mounted arm already…

Drake Moore, Matt Cheng, Xiang Zhi Tan +1

RoboticsĠun 10

TacForeSight Predicts Tactile Feedback From Force Data for Better Robot Handling

Yujie Zang, Yuhang Zheng, Xian Nie +7

AIĠun 9

AHA-WAM: Asynchronous World-Action Model for Faster, Smarter Robot Manipulation

Jisong Cai, Long Ling, Shiwei Chu +10

RoboticsĠun 9

How Robots Turn Actions Into Motion and Contact Images for Better World Models

An embodied world model that translates future actions into motion and contact images for video prediction on cloth manipulation tasks.

Zhenyu Wu, Xiuwei Xu, Yukun Zhou +8