Tag
The article announces oomwoo, an open-source, build-it-yourself robot vacuum cleaner that uses 2D LiDAR and ROS 2 for autonomous navigation, with no cloud required. It is designed for the maker community and is built in public from the start.
A paper proposing FLISP, a mapless path planning framework for cooperative UGV-UAV tunnel inspection using a single LiDAR-IMU suite, achieving 100% success rate and 7ms latency in a 1.2km tunnel.
OmniPath is a multi-modal agentic framework that combines OpenStreetMap network topology with aerial LiDAR data to audit wheelchair accessibility by analyzing physical barriers like slope and surface discontinuities at high resolution, validated against field surveys.
ShotcreteDepth is a bi-modal dataset of stereo RGB and LiDAR data from construction environments, designed to support research in depth perception under challenging conditions. The dataset includes 11,252 samples with 220 annotated, and is accompanied by a lightweight annotation tool.
KITScenes Multimodal is a high-fidelity European autonomous driving dataset with synchronized sensors, complete 3D HD maps, and four benchmarks for spatial learning and embodied AI research.
This week awesome-autoresearch added three items, including the autoslam project that applies Karpathy's autoresearch loop to LiDAR SLAM, and two blog posts analyzing the original experiments and revealing metric gaming.
Starbucks has discontinued its Automated Counting AI inventory system after 9 months due to inaccuracies like failing to distinguish between milk types, reverting to manual counting and a new daily replenishment model.
Sensor2Sensor uses diffusion models and 4D Gaussian Splatting to convert in-the-wild dashcam videos into multimodal autonomous vehicle logs (multi-view camera and LiDAR) for training and validation of autonomous driving systems.
A thread sharing a video of self-play RL training with lidar and PPO in Unity, followed by a lecture on building AlphaGo from scratch.
Ouster announces REV8, the first native color lidar sensor that fuses color and 3D data directly in silicon rather than in software, marking a hardware-level advancement in 3D sensing technology.
Hesai released Picasso, the world's first full-color LiDAR chip supporting up to 4,320 laser channels, achieving native pixel-level fusion of color perception and distance measurement at the hardware level. The chip will power Hesai's next-generation ETX series LiDAR sensors expected to enter mass production in the second half of 2026.