새로 나온 로봇, 피지컬 AI 논문과, 각 논문이 업계에 갖는 의미.
Extending single Unmanned Aerial Vehicles (UAVs) exploration methods to multi-UAV teams can improve coverage speed and robustness, but introduces challenges such as consistent mapping, safe navigation, and deployment strategy. In this work, we present a centralized multi-UAV exploration framework that enables the use of existing single-UAV sampling-based planners in a multi-UAV setting.
Mobile robots and vehicles carry synchronized multi-camera rigs, yet many streaming 3D foundation models are designed for monocular input, leaving efficient use of rig geometry a challenge. We present StreamRig, a freeze-and-stream framework that builds causal streaming odometry for calibrated rigs on a frozen multi-view 3D foundation model.
Preprint. The authors introduce GroundAnything, a 4B-parameter visual grounding model that uses bidirectional diffusion with blockwise denoising for parallel spatial decoding instead of sequential autoregressive token generation. Its autoregressive variant, GroundAnything-VLM, reports 72.42% across 30 grounding benchmarks, ahead of GPT-6 Astra at 71.35%. This matters for latency-sensitive robotics and interactive vision systems that need fast, precise localization.
연구진은 로봇 팔이 쓰는 분말 분배 도구의 형태와 강화학습 제어 정책을 함께 최적화하는 Tool-Policy Co-Design 프레임워크를 제안했다. BOHB 기반 외부 루프와 SAC 기반 내부 루프로 도구 깊이, 폭, 림 스파이크 구조를 탐색하고, 유사한 기하 구조의 정책을 재사용해 탐색 속도를 높였다. 실물 로봇 실험에서 공동 설계한 도구는 표준 도구 대비 실제 계량 오차를 45% 줄였고, 이 논문은 동료 심사를 거치지 않은 프리프린트다.
Physical LiDAR attacks are often evaluated using fixed primitives and manually selected parameters, despite their strong dependence on surrounding traffic. We present TACTIC, a scene-aware framework that uses a multimodal large language model (MLLM) to coordinate state-adaptive roadside LiDAR attacks.
Pretrained robot policies provide useful action priors, but long-horizon manipulation still requires coordination between semantic reasoning and physical execution. Semantic reasoning operates at a coarser timescale than physical interaction, while episode-level failures provide limited guidance on which system component should be revised.
A vision-language-action (VLA) policy can finish a manipulation task while knocking over objects unrelated to it, so task success alone does not show that the policy is safe to deploy in clutter. We study how to keep a pretrained VLA policy clear of such hazards at run time without retraining it, which requires guarding more of the arm than the end effector, following the hazard as it moves, and sharing onboard compute with the policy.
Echolocating bats can navigate dark and cluttered spaces using echolocation. Over a decade ago, BatSLAM showed that a robot with a biomimetic binaural sonar can build a topological map of the environment, by recognizing places from the received acoustic signals.
Robot navigation in dynamic, human-centered environments requires socially-compliant decisions grounded in robust scene understanding. Recent Vision-Language Models (VLMs) exhibit promising capabilities such as object recognition, common-sense reasoning, and contextual understanding, capabilities that align with the nuanced requirements of social robot navigation.
FailBank는 런타임 CBF 교정 신호를 관찰 전용으로 수집해 VLA 정책을 LoRA로 반복 개선하는 프레임워크다. VLA-Arena 정적 장애물 평가에서 π0.5 백본 기준 평균 성공률을 66.0%에서 74.5%로 8.5%p 높이고 정책 유발 누적 비용을 35.6% 줄였다. 런타임 차폐 AEGIS와 비교하면 성공률을 25.4%p 높였으며, 이 연구는 프리프린트이고 실물 로봇 실험은 보고되지 않았다.
We present Game-Guided Skill Discovery (GGSD), a framework that uses self-play in games to discover motor skills that are directly playable by humans. Playable skills provide a compact abstraction for controlling embodied agents through a small set of learned behaviors rather than low-level actions.
The task of 3D assembly requires translating an understanding of parts and their relationships into precise spatial arrangements. Can pretrained general-purpose agents assemble objects through visual interaction without additional assembly-specific fine-tuning?
Robot brain models integrate vision, language, and robot state to generate actions for complex manipulation tasks. Most predict fixed-length action chunks that may span multiple task phases.
이 논문은 로봇 안전 요구사항을 해석 가능한 1차 논리 규칙으로 표현하고, 시각 입력에서 안전 관련 술어를 학습하는 NEUPRO를 제안한다. REASON이라는 첫 실물 로봇 해석 가능 안전 벤치마크에서 NEUPRO는 과제 평균 0.92의 안전 분류 정확도를 보였고, 텐서 기반 추론기 대비 학습 9.1배, 추론 1.6배 빠르며 메모리 사용이 14.3배 낮았다. 이 연구는 아직 동료 심사를 거치지 않은 프리프린트다.
Mastering robot manipulation skills via reinforcement learning (RL) remains largely sample-inefficient. The most common RL algorithms rely on random action sampling to discover new strategies, resulting in agents that allocate most of their training budget to motions in free space, away from the contacts from which manipulation skills emerge.
World action models (WAMs) are large embodied policies that jointly predict future video and the actions to execute, emitting a fixed-length action chunk per inference call. Such a policy allocates its computational budget uniformly in time, unable to execute for longer over free-space motion or to spend more inference on contact-rich manipulation, which limits the throughput a WAM can reach when served in the cloud.
Multi-Agent Path Finding (MAPF) aims to find collision-free paths for multiple agents in a shared environment. Classical MAPF assumes that all static obstacles are known in advance, but real-world environments can change unexpectedly due to fallen objects, spills, or other local disturbances.
Open concrete canals play a vital role in water transportation, serving as primary water infrastructure for millions of people across the Phoenix, Arizona, metro area. Over time, the concrete canals can experience a range of issues, including canal lining deformation, cracked concrete, and sediment buildup on the canal floor.
DiffWAM은 동결된 비디오 기초 모델의 중간 예측 표현을 첫 프레임 기하 정보와 결합해 연속 카메라 궤적으로 직접 변환하는 내비게이션 세계-행동 모델이다. DiffWAM-1000 1,000개 평가에서 궤적 RMSE 0.3492m, 엔드포인트 성공률 74.40%를 기록했고, DiffWAM-Flash는 Jetson AGX Thor에서 모델 파이프라인 지연 1.080초를 보였다. 이 연구는 arXiv 프리프린트이며, 미래 비디오 생성과 다중 프레임 복원 없이 예측 표현만으로 구조적 UAV 비행을 수행할 수 있음을 보였다.
Vision-language-action (VLA) models can exceed 90% success on in-distribution tasks and withstand nuisance changes that preserve the required action, yet fail under counterfactual changes that demand a different action. Aggregate robustness scores can therefore conceal a more specific failure, in which a policy responds to both language and vision yet does not combine them to select the action the task requires.
논문은 arXiv 로봇 피드에서 가져옵니다. 요약이 아직 작성되지 않은 논문은 초록의 첫 부분을 보여줍니다.