VISUAL POSITIONING

Where LiDAR Gives Up Camera Finds the Position

Navi-VPS2 expands the area available for autonomous navigation with camera-based cm-level positioning in LiDAR SLAM-weak environments such as large glass walls, repetitive structure corridors, and open spaces. It operates without additional marker infrastructure using a dedicated mapping robot and AI dual-layer map DB.

Navi-VPS2

Key Features

4 features

Camera-Based cm-Level Positioning

Distinguishes locations using color and texture information in repetitive structures where LiDAR cannot differentiate. Leverages the visual features of the site itself without additional QR/beacon infrastructure.

Learn more ↓

AI Dual-Layer Map DB

Global Feature (place recognition, candidate extraction) → Local Feature (precise 6DoF positioning): Coarse-to-Fine structure maintains both accuracy and real-time performance even as DB size grows.

Learn more ↓

Automatic Environment Change Zone Separation

Dynamically excludes feature points of changed areas from matching. Responds stably to environmental changes without map reconstruction.

Automatic LiDAR SLAM Switching/Fusion

Automatically switches to VPS or fuses two estimates when Navi-Core LiDAR SLAM reliability decreases. Achieves high availability through positioning redundancy.

Dedicated Mapping Robot: Covers Even Robot-Inaccessible Areas with Backpack Mode

A dedicated mapping robot equipped with 8 cameras and 2 LiDARs was developed in-house. The fully detachable sensor unit switches between driving robot mount mode and backpack wear mode. The same equipment can build a map DB covering even narrow spaces, stairs, and temporary work areas inaccessible to robots.

  • 8 cameras + 2 LiDARs omnidirectional high-resolution data collection
  • Detachable sensor unit: dual use as driving robot mount or backpack
  • DB construction even for robot-inaccessible areas with backpack mode
  • AI Global+Local Feature dual-layer map DB
Dedicated Mapping Robot: Covers Even Robot-Inaccessible Areas with Backpack Mode
Product Image

Covering LiDAR-Weak Environments: Glass Walls, Repetitive Corridors, Open Spaces

If zones not covered by LiDAR alone exist, those areas must be excluded from automation or require human substitute labor. Applying Navi-VPS2 allows autonomous driving to continue even in these zones with camera-based positioning, practically expanding the operational area and ROI of the automated system.

  • LiDAR-weak environments: glass walls, repetitive corridors, open spaces
  • Kidnap situations: immediate Re-localization via single-frame Global Feature
  • Positioning redundancy → 24/7 high-availability site coverage
  • Eliminates additional infrastructure costs like markers and beacons

Compatibility & Integration

A compact add-on design mounts directly on a wide range of robot platforms. Standard interfaces make it easy to connect with your existing infrastructure.

ROS 1
ROS 2

Contact Sales about Navi-VPS2

Analyze your site with our experts and design the right solution. We support you through every step, from consultation to on-site installation.

Contact Us