HKUST-Aerial-Robotics/VINS-Mono

概览

解决什么问题

Monocular visual-inertial odometry (VIO) for UAV state estimation without GPS, providing 6-DoF pose and velocity estimates from a single camera and IMU.

主要用例

• Onboard pose estimation for autonomous flight
• Sensor fusion benchmarking
• SLAM/VIO algorithm comparison

次要用例

• Ground robot localization
• AR/VR tracking reference
• Multi-sensor calibration validation

何时使用

考虑使用

• Monocular camera + IMU sensor suite available
• ROS 1 (Melodic/Noetic) or ROS 2 (Humble) middleware
• Need open-source, actively maintained VIO with loop closure
• GPL-3.0 license compatible with project

采用前需验证

• ? Scale observability requirements (monocular scale drift)
• ? Initialization motion constraints (requires sufficient excitation)
• ? Real-time performance on target compute (CPU-only)
• ? Calibration quality of camera-IMU extrinsics/intrinsics
• ? GPL-3.0 obligations for proprietary deployment

在技术栈中的位置

技术层 perception_localization

从这里开始

采用检查清单

• 需验证 Scale observability requirements (monocular scale drift)
• 需验证 Initialization motion constraints (requires sufficient excitation)
• 需验证 Real-time performance on target compute (CPU-only)
• 需验证 Calibration quality of camera-IMU extrinsics/intrinsics
• 需验证 GPL-3.0 obligations for proprietary deployment

以上检查项只有在官方来源确认后才能标记为“已验证”；无法确认的保持未验证。

已知限制与未知项

已知限制

• Monocular scale ambiguity without external constraints
• Initialization fails with pure rotation or low texture
• CPU-intensive optimization; may need desktop-class compute
• ROS 1 primary; ROS 2 port community-maintained
• No built-in dense mapping or semantic understanding

替代与相关工具

替代方案

• open-vins — 替代
• kimera-vio — 替代

如何使用？

从下方记录的入口开始，然后对照技术清单进行验证。

• 官方网站 https://github.com/HKUST-Aerial-Robotics/VINS-Mono
• 仓库 https://github.com/HKUST-Aerial-Robotics/VINS-Mono
• 文档 https://github.com/HKUST-Aerial-Robotics/VINS-Mono

技术清单

• 通过 许可证已识别 已记录: GPL-3.0
• 通过 维护信号 活跃维护
• 通过 验证状态 官方确认
• 通过 已附加来源证据 1 个来源记录
• 需复核 已记录最新版本 未记录

官方链接

元数据与治理

数据集事实

以上事实仅来自官方数据集卡，未确认字段保持未知。

相关资源与依赖

• open-vins — alternative to (verified)
• kimera-vio — alternative to (verified)

近期动态

相关知识

下方相关指南/合集/对比的标题与摘要暂保留英文，待编辑翻译复核；事实字段与链接已验证且保持不变。

指南

• Evaluating SLAM / VIO Systems for UAV Use — A repeatable checklist for comparing open-source SLAM and visual-inertial odometry systems before integrating them into a UAV stack.

合集

• SLAM 与视觉惯性里程计技术栈评估 — 用于无人机导航、巡检与建图任务的开放源码定位与建图系统。

对比

• ORB-SLAM3 vs VINS-Mono for UAV State Estimation — Metadata and repository facts recorded from the official GitHub API on 2026-08-14. Sensor assumptions are recorded from official README statements. Flight-test behavior, benchmark numbers and certification are deliberately not compared because the indexed sources do not provide field-level evidence.