lidar
PulseAugur coverage of lidar — every cluster mentioning lidar across labs, papers, and developer communities, ranked by signal.
- 2026-05-18 research_milestone A new method for imaging hidden objects using consumer LiDAR was published. 来源
11 天有情绪数据
-
Waymo 自动驾驶出租车在洪水道路上遇到困难，导致服务暂停和召回
Waymo 的自动驾驶汽车在洪水道路上遇到了重大问题，导致服务中断和软件更新召回。这些自动驾驶汽车曾驶入洪水区并被困住，其中一起事件涉及车辆被冲走。鉴于 Waymo 依赖激光雷达和详细地图，理论上这应该有助于检测和导航此类危险，但这种情况令人惊讶。不过，该公司拒绝就这些故障的具体原因发表评论。
-
新方法利用证据图精炼激光雷达-相机标定
研究人员开发了一种新的激光雷达和相机系统标定方法，特别适用于农业环境。该方法采用“支持图驱动”技术来识别对准确标定最关键的观测，过滤掉噪声或模糊的数据。通过聚合对齐观测的一致性，该方法突出了可靠的标定证据，提高了在KITTI等数据集上的准确性。
-
新框架从路侧传感器生成车辆LiDAR数据
研究人员开发了RS2AD-LiDAR，一个旨在从路侧传感器观测生成车载LiDAR数据的新框架。该方法旨在克服传统自动驾驶系统单车数据采集的高成本和数据限制。该框架重建路侧LiDAR点云，合成高保真车辆数据，并在使用生成数据进行训练时，已证明了目标检测精度的提高。
-
新AI模型使用4D雷达进行可靠的人员检测
研究人员开发了一种名为TMVA4D的新型人工神经网络架构，用于使用4D雷达数据进行语义分割。该系统旨在提高自动驾驶汽车和机器人检测人员的可靠性，特别是在传统传感器（如摄像头和激光雷达）可能失效的严峻环境条件下。TMVA4D模型利用CNN和ConvLSTM编码器处理包括多普勒速度在内的4D雷达点云，并在区分人员与背景噪声方面显示出有希望的结果，即使在低能见度场景下也是如此。
-
STELLAR模型通过3D数据融合推进自动驾驶感知
研究人员开发了STELLAR，一款用于自动驾驶3D感知的新型大型模型，通过扩展稀疏窗口Transformer来整合激光雷达、雷达、摄像头和地图数据。该模型在包含5000万个驾驶示例和多达5亿个参数的数据集上进行训练，并在Waymo Open Dataset上达到了新的最先进水平。研究表明，通过大型数据集和计算能力扩展模型是推进自动驾驶感知系统的可行途径。
-
Sensor2Sensor converts dashcam video to AV sensor data
Researchers have developed Sensor2Sensor, a new generative modeling approach to convert in-the-wild dashcam videos into structured, multi-modal sensor data suitable for autonomous driving systems. This method addresses …
-
混合机器学习模型利用卫星数据改进森林高度估算
研究人员开发了一种混合机器学习模型，通过整合 TanDEM-X 和 Landsat 卫星的数据来改进森林高度的估算。该增强模型结合了光学 Landsat 数据，提供了关于森林结构的补充信息，解决了早期仅依赖 TanDEM-X 干涉相干性的模型中存在的歧义。在加蓬的洛佩国家公园进行的验证表明，与原始方法相比，均方根误差 (RMSE) 降低了 13.5%，平均绝对误差 (MAE) 降低了 16.6%，误差显著减少。
-
新方法利用3D和2D AI估计小麦穗体积
研究人员开发了一种新颖的混合方法，结合3D重建和知识蒸馏技术来估计小麦穗体积。该方法旨在克服传统测量方法的挑战，这些方法要么计算成本高昂，要么对环境条件敏感。通过将3D模型中的知识蒸馏到基于2D图像的Transformer中，该系统显著降低了平均绝对误差和推理时间，使其适用于高通量田间表型分析。
-
StruMPL model tackles forest biomass estimation with novel regression techniques
Researchers have developed StruMPL, a novel multi-task dense regression model designed to estimate forest aboveground biomass (AGB) using disparate data sources. The model integrates satellite lidar data, which provides…
-
LiDAR motion capture uses Bézier curves for improved accuracy
Researchers have developed a new framework called BMLiCap for more accurate 3D human motion capture using LiDAR data. This method employs Bézier curves to represent motion, which helps in creating a more coherent and le…
-
Robots use RGB cameras for 3D scene mapping, improving exploration
Researchers have developed a new framework for 3D scene graph generation that can operate using only RGB cameras, eliminating the need for depth sensors like LiDAR. This approach allows for more flexible deployment on v…
-
New learning-based method enhances LiDAR point cloud compression
Researchers have developed Inter-LPCM, a novel learning-based method for compressing LiDAR point cloud data. This approach improves upon existing techniques by utilizing inter-frame prediction to capture complex motion …
-
Consumer LiDAR can now image hidden objects using motion-induced sampling
Researchers have developed a new method for imaging objects that are not directly in a device's line of sight, utilizing consumer-grade LiDAR sensors. This technique, called Motion Induced Sampling, overcomes limitation…
-
SAGE3D模型通过新颖的注意力机制增强3D LiDAR角点检测
研究人员推出SAGE3D，这是一种新颖的基于Transformer的模型，用于检测LiDAR数据的3D点云中的角点。该模型采用分层编码器-解码器架构，并包含两项关键创新：软引导注意力（Soft-Guided Attention），在训练过程中利用地面真实标签来优化注意力；以及激励图神经网络（Excitatory Graph Neural Network），通过正向消息传递来提升高置信度角点预测。这种混合方法旨在提高多尺度角点检测的精度和召回率。
-
TriBand-BEV system achieves real-time LiDAR-only pedestrian detection
Researchers have developed TriBand-BEV, a novel real-time 3D pedestrian detection system using only LiDAR data. The system encodes 3D LiDAR point clouds into a lightweight 2D Bird's Eye View (BEV) tensor with three heig…
-
MTA-RL framework enhances urban driving with multi-modal AI
Researchers have developed MTA-RL, a novel framework that integrates multi-modal transformer-based 3D affordances with reinforcement learning for robust urban autonomous driving. This approach fuses RGB images and LiDAR…
-
AI research advances autonomous driving perception and safety
Researchers are developing advanced AI techniques to improve autonomous driving systems. One approach, CaAD, focuses on causality-aware end-to-end modeling to better predict vehicle and agent interactions, showing stron…
-
South Korea and MicroVision/Avular advance AI-powered drones for defense
South Korea's National Intelligence Service and military are pooling their data to counter drone threats, integrating AI and real-world training in response to the conflict in Ukraine. In parallel, MicroVision and Avula…
-
GEM model generates LiDAR world models for autonomous driving
Researchers have developed GEM, a generative LiDAR world model designed to simulate environmental dynamics for autonomous driving. The model utilizes a deformable Mamba architecture to overcome challenges with disordere…
-
AITO M6 integrates advanced Huawei ADS 4.1 and air suspension, offering premium features at a competitive price.
The AITO M6 sedan features advanced Huawei ADS 4.1 intelligent driving system with a high-spec 896-line lidar and 36 sensors across all trims. It also boasts a premium chassis with standard air suspension and CDC damper…