gps-denied 话题在 GitHub 上没有填写简介，这里按 Star 数汇总最相关的热门仓库。
Autonomous UAV navigation in GNSS-denied indoor spaces with no prior map: stereo VIO (MSCKF) and lidar-inertial odometry with map-based relocalization (SLAM), 3D occupancy mapping, D* Lite path planning and GPU MPPI obstacle avoidance. PX4 + ROS 2 + Gazebo SITL, DARPA SubT Urban.
GPS-denied drone navigation layer: optical flow + offline map geo-match + EKF → MAVLink (sim-validated).
The framework for visual place recognition in outdoor environments using image sequences.
GPS-Denied Indoor Navigation System for Drones Autonomous drone navigation indoors without GPS using optical flow, IMU, and lidar sensor fusion. Features real-time SLAM, obstacle avoidance, and waypoint missions. Built with PX4/ArduPilot, ROS2, and MAVSDK-Python on Raspberry Pi.
GPS-denied autonomous UAV swarm platform with edge AI, distributed swarm intelligence, VIO/EKF fusion, UWB/TDOA localization, secure telemetry, CBOR-based edge communication, and post-quantum security roadmap.
The estimated navigation solution is published to ArduPilot through the MAVLink External Navigation interface (VISION_POSITION_ESTIMATE), allowing EKF3 to fuse the companion computer's state estimate.
INS System for UAV Navigation (GPS-Denied) A high-fidelity MATLAB implementation of a strapdown Inertial Navigation System (INS) for UAVs operating in GPS-denied environments, leveraging an Extended Kalman Filter (EKF) for multi-sensor fusion.
Information and relevant research from 2021 Auburn University REU on Smart UAVs. Computer Vision through terrain image processing, specifically through implementations of SIFT (feature matching) and template matching.
Robust UAV Path Planning Using RSS in GPS-Denied and Dense Environments Based on Deep Reinforcement Learning
🛸 Navigate urban environments with this autonomous UAV system using real-time occupancy grid mapping and dynamic obstacle avoidance for safe and efficient flights.
Perform real-time GPS-denied INS navigation for drones using a Pixhawk Cube Orange and Raspberry Pi 4 via MAVLink.
Drift-free vision-only autonomous hover for UAVs by getting motion_delta from the image sensor, baro and IMU (5–900 m tested in PX4 SITL, no height ceiling found in sim).
VIO + Footstep Planner for GPS-denied quadruped navigation - 98% success rate, 60% drift reduction, real-time autonomous navigation with loop closure
Estimating a fixed-wing UAV's latitude, longitude and altitude from onboard sensors alone when GPS is unavailable. Neural dead reckoning in PyTorch.
Access NavAbility(TM) Accelerator features from JuliaLang.
GPS-free iPhone driving prototype: speed from the road-bump echo between the axles, position from turn matching on offline maps (Kyiv, Lviv region). Noncommercial.
GPS-denied pedestrian dead reckoning: learned inertial velocity fused in a 2D error-state Kalman filter. SIH 2026 · SIH26168 · ISRO.
Selected code from my M.Sc. thesis: Cyber-resilient multi-modal sensor fusion for autonomous ship navigation
Turn a fresh NVIDIA Jetson Orin Nano into a flight-ready drone companion computer. CUDA, TensorRT, PyTorch, MAVLink, SSH hardening, and more.
Source code for virtual-based experiments presented in the paper: UAV Navigation Using EKF-MonoSLAM Aided by Range- to-Base Measurements