robotics
PulseAugur coverage of robotics — every cluster mentioning robotics across labs, papers, and developer communities, ranked by signal.
14 天有情绪数据
-
韩国与卡塔尔探讨人工智能和机器人投资伙伴关系
韩国与卡塔尔正在探索在半导体和人工智能等先进产业领域加强投资合作。两国旨在结合韩国的制造能力和卡塔尔的资本，以促进战略伙伴关系。讨论还涉及在制造业和机器人领域的人工智能转型等新兴行业的潜在合作，以及韩国生物技术和AI芯片公司的投资机会。
-
New benchmark and multi-agent framework boost physics-aware simulation accuracy
Researchers have introduced PhysCodeBench, a new benchmark designed to evaluate the ability of AI models to perform physics-aware symbolic simulation of 3D scenes. This benchmark includes 700 manually created samples co…
-
MARL enables robots to cooperatively monitor indoor environments
Researchers have developed a new multi-agent reinforcement learning framework for robots to cooperatively monitor indoor environments. This approach optimizes robot movement to directly enhance monitoring accuracy, unli…
-
New methods KERV and HeiSD accelerate embodied VLA models with kinematic awareness
Two new research papers introduce methods to accelerate the inference speed of Vision-Language-Action (VLA) models used for robot control. KERV utilizes a Kalman Filter to predict actions and adjust acceptance threshold…
-
Researchers use SHAP and RL to improve robot generalization and affordance reasoning
Researchers have developed a framework using SHapley Additive exPlanations (SHAP) to analyze and improve the generalizability of reinforcement learning (RL) algorithms in robotics. This approach quantifies the impact of…
-
New flow-matching models offer advanced control for generative tasks
Researchers have developed two novel approaches to enhance flow-matching generative models. One method, HardFlow, reframes hard-constrained sampling as a trajectory optimization problem, allowing precise constraint sati…
-
Libra-VLA model balances learning for enhanced robotic manipulation
Researchers have introduced Libra-VLA, a novel Vision-Language-Action (VLA) model designed for robotic manipulation. This architecture employs a coarse-to-fine dual-system approach, decoupling the learning process into …
-
FeudalNav框架简化了机器人的视觉导航
研究人员推出了一种名为FeudalNav的新型分层框架，专为机器人视觉导航而设计。该系统将导航分解为多个级别，利用一个简单的航点选择网络和一个由视觉相似性组织的独特潜在空间记忆模块。FeudalNav旨在无需依赖传统的度量地图或里程计即可导航未知环境，并在Habitat AI模拟中取得了有竞争力的结果。该框架还支持交互式导航，表明最少的人工干预可以显著提高性能。
-
AI robots now beat elite table tennis players, showcasing advanced robotics
Researchers have developed an AI-powered robot capable of competing at a high level in table tennis. This robot has demonstrated a significant win rate against human players, even those considered elite. The system's ab…
-
Montréal's AI and robotics talent powers discreetly in space missions
Montréal is playing a significant, though often unseen, role in the space industry, contributing robots, AI expertise, and local talent to various missions. This contribution spans across different aspects of space expl…
-
中国将于2026年举办多场人工智能和智能设备贸易展览会
2026年中国将举办多场专注于人工智能和智能设备贸易展览会。这些活动旨在将企业与人工智能解决方案、机器人技术和数字化转型服务联系起来。主要展览包括广州国际智能装备暨人工智能产业博览会、上海科技周、恩摩尔AI驱动工业大会暨展览会以及上海国际安博会。
-
LeHome simulation environment advances household robotics with deformable object manipulation
Researchers have introduced LeHome, a new simulation environment specifically designed for robotic manipulation of deformable objects in household settings. This environment addresses the limitations of existing simulat…
-
New system improves camera-to-mocap calibration for AR/VR and robotics
Researchers have developed a new system for calibrating and verifying multi-camera setups with optical motion capture, specifically addressing challenges posed by fisheye lenses. The system enhances robustness against c…
-
NVIDIA's Jim Fan: VLA, Teleoperation Obsolete by 2026 in Robotics
NVIDIA's Jim Fan predicts a major shift in robotics, declaring VLA architectures and teleoperation obsolete by 2026. He advocates for a new era driven by physics-simulating world action models and autonomous data collec…