New robotics and physical-AI papers, with what each one means for the industry.
When interacting with an unfamiliar deformable material, a robot lacks prior knowledge of its physical properties and how it will respond to applied forces and motion. Rapid online identification is therefore essential for reliable manipulation.
Robotic foundation models (RFMs) are increasingly capable of general-purpose manipulation, yet reliable physical interaction remains challenging in contact-rich settings. We present FP2, a lightweight downstream interface that equips task-adapted RFMs with explicit force control while preserving their action-generation capability.
Simulated manipulation benchmarks are the standard tool for evaluating vision-language-action (VLA) policies and the acceleration methods that reduce their inference latency for on-robot deployment. On these benchmarks, we observe that some training-free acceleration methods, which approximate the baseline policy's computation, achieve higher measured success rates than the baseline itself.
World-action models (WAMs) jointly predict how a scene will evolve and how an agent should act, however joint generation alone does not necessarily impose a shared geometric constraint on these predictions. We present PhysWAM, a unified world-action model for autonomous driving that co-denoises multiview video, metric depth, and ego motion within a single flow-matching transformer.
Test-time adaptation for vision-language navigation (TTA-VLN) enables pretrained policies to adapt online to unseen environments using only test-time observations and interaction history. However, distribution shifts can distort local action preferences and lead to off-course decisions.
Improving robot policies on new tasks without collecting additional expert demonstrations remains a central challenge in robot learning. World action models (WAMs) use broad video priors to jointly predict future videos and actions, offering a potential source of supervision for adapting to new tasks.
This preprint introduces DROM, a language-guided diffusion policy that learns 17 manipulation skills from only 1–4 human demonstrations per skill, using Dynamic Movement Primitives to expand each into 300 synthetic trajectories. A single temporal U-Net with cross-attention generates skill-consistent motions on Franka Panda and FANUC CRX25ia arms, and long-horizon tasks are handled by an LLM that sequences skills; end-to-end task success ranges from 60.9% to 86.0% depending on task. It matters for industrial robotics because it cuts manual data collection and enables natural-language task programming across multiple hardware platforms.
Vision-language-action (VLA) models typically operate on RGB images produced by a fixed camera image signal processor (ISP), leaving the imaging pipeline outside the learning and evaluation loop. We systematically examine the consequences of this overlooked design choice across five fundamental ISP dimensions: gain, sensor noise, chromatic response, tonal response, and bit depth.
Research in modular robotics has produced capable approaches allowing a robot's morphology to change online, with recent efforts also developing approaches to decide which morphology to assume and automatically propagate that decision into the robot's motion planning and control. These approaches are powerful and increase adaptability in the field.
General-purpose physical AI models must combine broad visual and linguistic capabilities with precise control across robot embodiments and efficient adaptation to downstream tasks. We introduce Rho, a family of open-weights VLA models for bimanual manipulation designed for data-light task adaptation on 3 embodiments representative of dual-arm robots across research labs and the industry -- YAM Box, UR AI Trainer, and FR3 Duo.
Co-training a VLA on heterogeneous data (multi-robot, web, high-level subtask labels) lets a mobile manipulator clean unseen homes end-to-end.
A slow VLM planner (7–9 Hz) and a fast visuomotor policy (200 Hz) control a full humanoid upper body from language.
RL policies trained in GPU simulation transfer zero-shot to a commodity humanoid on rough terrain.
Survey of sensor modalities, coverage and learning methods for in-hand manipulation.
A video world model trained on robot egocentric footage predicts outcomes of actions for evaluation and planning.
Cobot adoption in Korean SMEs is limited by integration cost more than by unit price.
Papers come from arXiv robotics feeds. Where our summary is not written yet, you see the opening lines of the abstract.