[Edaily Reporter Shin Ha-yeon ] SOS LAB Co., Ltd.(464080), a company specializing in LiDAR-based spatial intelligence solutions, announced on the 29th that it has begun technology development and field demonstrations in collaboration with the Korea Electronics Technology Institute (KETI) and Roas to commercialize an unmanned perimeter security and patrol platform based on quadruped robots.
This project, titled “Development and Demonstration of Fusion Technology Combining Multimodal Sensor-Based Spatial Intelligence and Quadruped Robots for the Establishment of an Unmanned Surveillance and Patrol Platform,” aims to advance technology currently at Technology Readiness Level (TRL) 6 to Level 8, making it suitable for use in real-world operational environments.

SOS LAB Co., Ltd. is developing a multimodal sensor—which integrates 3D LiDAR, an RGB camera, a thermal imaging camera, and an inertial measurement unit (IMU)—along with a spatial intelligence platform. Through this, the company plans to recognize and assess abnormal situations—such as people, objects, fires, and intrusions—in real time, and implement natural language reporting capabilities using a large language model (LLM)-based control agent.
In this project, SOS LAB Co., Ltd. will be responsible for integrating multimodal sensors, developing spatial intelligence algorithms, and building the control system. The Korea Electronics Technology Research Institute will handle the development of disaster and safety AI models and system validation, while ROAS will be in charge of developing the quadruped robot platform and system integration technology.
The platform being developed is designed with an open architecture that is not tied to any specific robot; it will be modularized so that it can be applied not only to quadruped robots but also to wheeled robots, bipedal robots, and fixed surveillance systems.
Jeong Jong-kyu, Head of the Innovation Vision Center at SOS LAB Co., Ltd., said, “We will integrate domestically developed sensors, spatial intelligence technology, and a physical AI platform to realize a ‘robot brain platform’ capable of situational awareness, decision-making, and response across various robots and field environments.”