새로 나온 로봇, 피지컬 AI 논문과, 각 논문이 업계에 갖는 의미.
Encore는 몇 개의 시연을 키프레임과 그리퍼 이벤트 같은 증거 패킷으로 압축하고 코딩 에이전트가 정책 프로그램을 시뮬레이션과 실물 양팔 로봇에서 반복 수정하게 하는 프레임워크다. LIBERO-PRO의 60개 변형 셀에서 시연 3개를 쓴 프로그램이 96.3%, 시연 없이 쓴 프로그램이 94.9%의 성공률을 보였고, 실물 로봇 두 과제에서는 각각 9/10을 기록했다. 아직 동료 심사를 거치지 않은 프리프린트다.
이 프리프린트 논문은 로봇이 명령한 동작과 실제 실행한 동작의 잔차를 자기지도 신호로 사용해 VLA 정책을 배치 중 온라인으로 적응시키는 self-compensating VLA를 제안했다. 저자들은 마찰, 백래시, 컴플라이언스, 중력 보상 오차를 결합한 시뮬레이션 벤치마크 RoboStress와 두 대의 실물 Piper 팔에서 평가했고, 두 팔 각각에서 평균 작업 성공률을 30%포인트 이상 끌어올렸다고 밝혔다. 이 연구는 아직 동료 심사를 거치지 않았다.
Embodied systems must make knowledge acquired during one encounter usable in another despite changes in viewpoint, motion, and illumination. Yet aggregate cross-video accuracy conflates failures of local perception with failures to preserve observation identity, establish correspondence, and compose evidence, obscuring whether local video understanding actually transfers.
We propose a QCQP-representable IMU pre-integration factor that enables certifiable estimation with pre-integrated inertial measurements. To the best of our knowledge, this is the first work to directly incorporate IMU pre-integration into certifiable estimation.
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.
Large language models (LLMs) are increasingly used as high-level planners in robot navigation, but their outputs may become unreliable when instructions are ambiguous, unsupported by the environment, or semantically inconsistent. This paper presents a Risk-Aware Semantic Grounding framework for trustworthy LLM-based robot planning.
In this preprint, the authors build an onboard perception pipeline that lets a lightweight Explorer robot find minerals in underground mines from free-text prompts and hand compact, oriented 3D inspection targets to a heavier Inspector robot. Field tests in a test mine and an active magnesite mine detected two mineral types without retraining and condensed about 38,000 raw mineral points into a few dozen polygons, roughly two orders of magnitude smaller. This matters because autonomous mining inspection requires no mine-specific labels and low-bandwidth robot-to-robot handover.
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.
Grinding and sanding are fundamental processes in industrial robotic surface finishing. However, physical trials are expensive and consume workpieces, making reproducible experiments difficult.
Robots deployed in the physical world must be able to improve beyond their initial training as they encounter new situations and failures. For this improvement to scale across tasks, it must make effective use of experience without requiring human demonstration of each correction.
Diffusion and flow-matching Vision-Language-Action (VLA) policies generate action chunks through iterative denoising, incurring substantial inference latency that severely limits real-time robotic control. Existing acceleration methods treat an action chunk as a monolithic computational unit, ignoring a crucial physical reality of receding-horizon control: actions are generated jointly but consumed sequentially, resulting in inherently heterogeneous execution urgencies.
Video-based policy learning is particularly promising, as it illustrates target behaviors without requiring action annotations or embodiment-matched demonstrations. A central challenge is deciding what information should be transferred from the video to the robot.
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.
DROM은 DMP 기반 데이터 증강과 언어 조건 확산 모델을 결합해, 적은 수의 인간 시연으로 17개 조작 프리미티브를 하나의 생성 정책으로 학습하고 대규모 언어 모델로 장기 작업을 분해해 실행하는 프레임워크다. MuJoCo 시뮬레이션과 Franka Emika Panda, FANUC CRX25ia 실물 로봇 실험에서 다중 기술 실행과 장기 작업 구성 전반에 걸쳐 MPD와 Behavior Cloning 대비 우수한 성공률을 보였다고 저자들은 보고했다. 이 연구는 아직 동료 심사를 거치지 않은 프리프린트다.
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.
논문은 arXiv 로봇 피드에서 가져옵니다. 요약이 아직 작성되지 않은 논문은 초록의 첫 부분을 보여줍니다.