Robotics Research Hub
Latest cs.RO robotics preprints from arXiv plus ZeroGantry Frontier analysis of academic and lab developments — foundation models, sim-to-real, manipulation, and locomotion research that matters for builders.
Latest articles · Patent Watch · Methodology
Frontier coverage
- NVIDIA GR00T 2 Hits 95% Success With Five Demos — NVIDIA unveiled GR00T 2, its world action model pretrained on hundreds of thousands of hours of human video and robot data. The model reaches 95% average succes
- Boston Dynamics Atlas GlobalFoundries Pilot Begins Early 2027 — Boston Dynamics and GlobalFoundries announced on November 17 that Atlas humanoid testing will start in early 2027 at semiconductor fabs. GlobalFoundries becomes
Recent arXiv cs.RO papers
- $Z$-sight: Unifying Vision-Language-Action and Latent World Modeling — Junjie Ye, Yiqi Zhao et al. — We introduce $Z$-sight, a unified robot foundation model that couples vision-language-action (VLA) and generative latent world modeling. Video-based world actio
- Phoenix: Learning Scene Structure Representations for Generalizable World-Action Models — Yupeng Zheng, Chaoyue Li et al. — World-action models jointly predict future observations and actions, yet reconstruction-oriented visual representations do not explicitly emphasize the spatial
- Grana: Compositional Reward Modeling for Long-Horizon Robot Manipulation — Zi-ang Cao, Kaiyuan Zheng et al. — Robot reward modeling faces two challenges: generalizing across tasks and accurately tracking progress in long-horizon manipulation. General-purpose reward mode
- Render the Robot, Generate the World: One Video World Model Across Embodiments — Yash Jangir, Karan Mirakhor et al. — Predicting the consequences of an agent's actions requires modeling both the body's motion and the environment's response. Most video world models attempt to le
- Teleoperated Haircutting with Delay Compensation and Passivity-Based Control for Unilateral Internet Teleoperation under Real Communication Uncertainties — Viridiana Luna-Zempoalteca, Zhendai Huang et al. — Robotic haircutting requires accurate motion execution and reliable communication to safely reproduce the operator's commands. Internet-based teleoperation is a
- Learning to Interact in Multi-Agent Self-Play for Controllable Autonomous Driving Scenario Generation — Beyazit Yalcinkaya, Ethan Pronovost et al. — Large-scale, multi-agent self-play trains robust driving policies effectively in simulation, but controlling these models to generate specific, safety-critical
- NereidVLA: Learning to Understand and Navigate the Underwater World — Chenjie Zhao, Zhiliang Zhu et al. — Underwater navigation requires translating visual spatial cues into motion, yet marine vision-language datasets rarely include paired robot trajectories. We pre
- Communications-free Collective Intelligence Via Allocentric Belief — Cameron Mehlman, Arjun Mulchandani et al. — A team of skilled card players share no information amongst each other, and yet still are able to coordinate their play by watching the same players and reading
- Does the Action Matter? Probing Action-Conditioned Semantic Forecasting as a World Model for Autonomous Driving — Binay Kumar Singh, Niels Da Vitoria Lobo — World models are increasingly used in autonomous driving to predict how a scene evolves under the ego vehicle's actions, supporting counterfactual reasoning and
- DiWALI: Diffusion Policy Learning in World Models via Latent Action Inference from Human and Robot Play — Akshay L Chandra, Iman Nematollahi et al. — Learning robot manipulation is constrained by the availability of action-labeled robot data, for both imitation and world models. Humans can demonstrate skills
- RoboMorpher: Scaling Robot Morphology Representations for Cross-Embodiment Action Prediction — Wenwei Lin, Yan Qin et al. — Learning shared action representations is central to cross-embodiment manipulation. However, identical joint values can produce different spatial configurations
- Planner-Controller Interaction for Ackermann Navigation: A Matched Simulation and Hardware Benchmark — Moaiz Saeed, Marah Saleh et al. — Global planners and path-tracking controllers are often evaluated independently, although their interaction determines how a vehicle follows a planned path. Whe
- HINT-WAM: Hindsight-Informed Training for World Action Models — Phuong Tuan Dat, Qi Li, Xinchao Wang — World action models (WAMs) have shown promising generalization and data efficiency in robot manipulation by coupling action generation with predictions of futur
- Counterfactual World Models: JEPAs That Reason About "What If" — Randall Balestriero, Yann LeCun, Kerda Varaku — World models operate in a world they cannot fully observe. When a hidden disturbance, such as a current pushing a boat, changes the outcome of an action, a coun
- Hierarchical Admittance-Impedance Control for Mobile Manipulator Load Transfer — Minghao Xu, Thushara Sandakalum et al. — This paper presents a hierarchical admittance-impedance controller that combines an inner Cartesian impedance controller, an outer 6D admittance supervisor, and
- DRIFT: Detachable Reasoning for Runtime-Flexible and Efficient End-to-End Driving — Emad Aghajanzadeh, Bethany Anne Wong et al. — The integration of vision-language models (VLMs) into end-to-end (E2E) autonomous driving has demonstrated strong potential by providing scene description and r