Neue Studien zu Robotik und Physical AI, jeweils mit ihrer Bedeutung für die Branche.
Reliable real-time failure detection is critical for deploying learned robotic policies, yet defining failures explicitly is infeasible in open-world settings. This motivates failure detection from nominal demonstrations alone.
Scaling robot manipulation policies requires large amounts of diverse, action-labeled data, yet collecting such data in the real world remains expensive and difficult to scale. Existing methods either depend on human teleoperation, rely on simulation with limited visual realism, or use human videos that lack embodiment-specific annotations.
Automating biological laboratories requires robots to perform dexterous tool interactions under realistic visual conditions while handling dynamic experimental phenomena, presenting significant challenges for robotic manipulation. To enable systematic evaluation under these conditions, we introduce AutoBio+, a simulation benchmark and framework for dexterous manipulation in realistic biological laboratory environments.
World-action models (WAMs) predict a robot's actions together with the observations those actions will produce. However, the strongest models rely on video diffusion transformers that are far too computationally expensive for resource-constrained edge devices.
Existing real-world reinforcement learning methods improve robot policies by combining human corrections with autonomous rollouts. However, accumulated experience varies in its value for policy improvement, and even successful trajectories can contain inefficient or redundant behavior, motivating more selective experience reuse.
This paper presents a dual-quadrotor hook- suspended transport system for cooperative aerial manipulation of a payload that is not rigidly attached to a fixed point along the cable. Two quadrotors are connected by a single inextensible cable that passes through a frictionless hook mounted on the payload, allowing the cable to slide between its two segments.
The NIST Assembly Task Board #1 remains a challenging benchmark for learning-based methods in robotic manipulation. Over short horizons, it demands precise, contact-rich control of small, tight-fitting parts.
Autoregressive robot policies rely on action tokenizers to convert continuous action chunks into compact sequences of discrete prediction targets. Existing tokenizers can reconstruct action chunks from discrete codes, but often leave implicit what each individual token contributes to the reconstructed action.
Collecting dexterous manipulation demonstrations for each robot hand requires repeated task-specific data acquisition. We ask whether one shared corpus of natural human demonstrations can instead serve as a reusable data interface for different dexterous hands, without collecting task demonstrations on each target robot.
Action-conditioned world models offer a promising foundation for generalist robotic intelligence by predicting the consequences of actions. However, existing robotic datasets provide limited coverage of the diverse interaction dynamics required for generalizable world modeling.
World Action Models (WAMs) inherit powerful priors from pretrained video generation models, but their variational autoencoder (VAE) representations are optimized for visual reconstruction rather than manipulation. External visual representations can improve manipulation, but existing methods couple representation choices with different integration mechanisms, obscuring their individual effects.
PEWAM is a preprint describing a 71.5M-parameter world-action model that compiles robot demonstrations into editable “effect programs”—3D object keypoint trajectories, grasp contact points, and final scene configuration—rather than actor-specific motions. On held-out LIBERO-Goal tasks, one program from a single demonstration completes 40 of 90 episodes, while the same backbone under goal-image or language conditioning and prior demonstration-conditioned methods complete at most 19. This matters because turning demonstrations into coordinate-based programs enables editing and cross-robot-arm reuse, a potential route to more flexible robot programming.
Das Paper stellt DynaHarness vor, einen Preprint, der einen dynamischen physischen Harness beschreibt, der semantische Planung eines Qwen3-VL-4B-Slow-Brain mit physischer Governance durch einen gemeinsamen Ausführungskontrakt verbindet. Auf LIBERO-Pro erreicht das eingefrorene System 75,2 % auf 800 neu gesampelten Startzuständen, die eingefrorene VLA-Politik 17,5 %. Der Ansatz zeigt, wie Ausführungsevidenz gezielte, validierte Revisionen wiederverwendbarer Fähigkeiten ermöglicht.
Das Preprint stellt Dream4ACT vor, ein Video-Aktions-Weltmodell, das Ziel-Gelenkkonfigurationen als feste Multiview-Aktionsbilder (Action Views) repräsentiert und damit mehrere Roboterembodiments über eine gemeinsame Schnittstelle trainiert. In Simulation erreicht das Modell durchschnittlich 88,98 % Erfolg auf RoboTwin 2.0 und einen TriWorldBench-Score von 65,66; auf vier realen Plattformen liegt der mittlere Erfolg bei gemeinsamen Aufgaben zwischen 85,0 % und 90,0 %. Die trainierungsfreie Rückgewinnung ausführbarer Gelenkziele über URDF-Optimierung macht die Aktionsdarstellung direkt für Steuerung nutzbar, ohne gelernte roboterspezifische Decoder.
World-Action Models (WAMs) couple visual dynamics with action prediction, bringing the rich priors of pretrained video models to robotic manipulation. However, their multi-view interfaces typically tile images or concatenate tokens, leaving the geometric relationships among synchronized cameras implicit.
Robot policies are frequently trained from human corrections, yet teleoperating a robot to provide corrections is burdensome, and human demonstrators are not always optimal. We propose Blended DAgger (BlenDAgger), an approach for collecting data to train imitation learning policies by using shared control to blend the policy's and demonstrator's actions during interventions.
Real-world robot learning is constrained by the cost of collecting experience and evaluating candidate behaviors. Video generation models offer a scalable foundation for visual simulators that predict action outcomes before physical execution.
We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations. Despite its growing promise, the problem itself remains under-defined: a visual demonstration simultaneously conveys action trajectories, object semantics, manipulation affordances, spatial relations, and task goals, making it unclear what information the robot is actually expected to follow.
While contact-rich manipulation requires deliberate regulation of interaction forces, recent approaches to robot manipulation learning predominantly represent actions as target positions or poses. Even methods that incorporate force sensing either use it solely as an observation or, when predicting forces as part of the output, rely on a hybrid force controller.
Vision-language models can coordinate long-horizon robot manipulation, yet successful task reasoning still depends on whether local physical interactions produce the intended effects. We study how repeated interaction can improve this capability without updating the base model.
Die Studien stammen aus den Robotik-Feeds von arXiv. Solange unsere Zusammenfassung fehlt, erscheinen die ersten Zeilen des Abstracts.