Daegu Gyeongbuk Institute of Science and Technology

Dae Han Hong

Tae Hun Kang

Jung Hyun Chu

Sung Gil Wee

One embodiment of the present invention provides an intersection point pattern recognition system using sensor data of a mobile robot, comprising: a mobile robot that autonomously drives by using sensor data received from a sensor unit and an intersection point pattern recognition model provided by a management server; and the management server that receives usage environment information of the mobile robot and generates the intersection point pattern recognition model of the mobile robot to provide the intersection pattern recognition model to the mobile robot, wherein the management server comprises: a map generation unit for receiving the usage environment information of the mobile robot and generating a route map of the mobile robot on the basis of the usage environment information; a normalization unit for generating a virtual map by normalizing the route map according to a preset rule; and a learning unit for generating the intersection point pattern recognition model by using the virtual map and the sensor data of the mobile robot as learning data.

Growing community of inventors

Daegu, South Korea

Dae Han Hong