AiR Lab logo Dongguk University
  • Home
  • Members
  • Research
  • Projects
  • Publication

Research

AiR Lab builds intelligence systems for mobile robots, spanning physical intelligence, 3D spatial perception, multi-robot coordination, and robot digital twins.

Conceptual robot world model connecting observations, prediction, and action
01

Physical AI

Predictive world models, vision-language-action models, and spatially grounded planning for robot behavior.

VLA & World Models 3D VLP
View details →
Spatial representations of objects and their relationships over time
02

Spatial AI

Visual localization, 3D reconstruction, active scanning, and object-level understanding of static and dynamic scenes.

Localization & SLAM MVS & Neural Rendering Active 3D Scanning 3D Scene Graphs Multi-Object Tracking
View details →
Multi-Robot System: Robot Fleet Management System overview
03

Multi-Robot System

Fleet coordination, online task allocation, multi-agent pathfinding, and language-guided planning with failure recovery.

Fleet Management System Decentralized MAPF Pickup and Delivery Language-Guided Planning
View details →
Robot Digital Twin: National Robot Test Field digital-twin environment
04

Robot Digital Twin

Layered information models, Real-to-Sim modeling, and interoperability testing for reusable robot simulations.

Layered Information Models Real-to-Sim Modeling Interoperability
View details →

Contact

Prof. Soohwan Song

Room #10108, New Engineering Building

College of AI Convergence, Dongguk University

Tel. 02-2260-3765

songsh@dongguk.edu

AI & Robotics Lab (AiR Lab)

Developing intelligence systems for mobile robots through 3D perception, path planning, and motion control.

© 2026 AI & Robotics Lab, Dongguk University. All rights reserved.