본문 바로가기 대메뉴 바로가기
open
Close

Artificial Intelligence Laboratory

Research

Present Research

Development of AI Technology to Generate and Validate the Task Plan for Assembling Furniture in the Real and Virtual Environment by Understanding the Unstructured Multi-modal Information from the Assembly Manual (2019~2020)


Main Objectives

Development of AI technology for figuring out the assembly process of furniture by understanding unstructured multimodal information in the assembly manual
Development of algorithms to generate and validate the task plan for assembling furniture in the real and virtual environment
Acceleration of activating the AI-Robot convergence research by adopting open software development method and MIT License

Main Functions

Automatic extraction of required images and text from assembly manual images
Inference of relationship between assembly process and parts by understanding images and texts in the assembly manual
Recognition of the component type and its region of interest on the image of simulation-real-manual environment
Comparison of similarity among simulation, real, and manual environment for analyzing the relationship among environments and real-time validation of assembly completion
Task plan generation by recognizing the combining regions and the combining relationship correspoding to each component
Real-time task validation for the failure response strategy

Development of Intelligent Meal Assistant Robot with Easy Installation for the Elderly and Disabled (2019~2021)

Development of Customized Food Dataset for Intelligent Meal Assistant Robot

Selection of 50 kinds of food suitable for intelligent meal assistant robot
Development of dataset for food detection system

Development of Food Type Recognition Algorithm

Development of deep neural network for Korean meal assistant robot

Development of Intelligent Meal History Management Service

Advancement of food type recognition algorithm
Development of food type recognition system for intelligent meal history management service
Development of intelligent meal history management service

Development of AI-based Shape Recognition Technology for Metalic Parts (2019)

Dataset Construction

DB server construction for the image analysis
CAD image DB construction about the feature shape
DB construction for the various arrangement of components

Development of Shape Recognition Deep Learning Model

Construction of deep learning system for the shape recognition on the CAD image
Deep learning model development based on the region proposal method (R-CNN) and single shot detector method (YOLO, SSD)

Development of Information Generation Module

Recognition of the ROI of the feature shape
Estimation of the type, location, and direction of feature shape
Estimation of the location of the feature shape

A Research on the Deep Learning based Speech Detection for Sound-Haptic Transformation (2019)

Main Objectives

Development of speech detection dataset for the online and the mobile phone game
Validation of the voice activity detection on the real-time game playing environment

Sim-to-Real Deep Reinforcement Learning for Visuomotor of Robots (2019~2021)

We are developing Sim-to Real transfer algorithm, which enables direct application of deep reinforcement learning model trained in a simulated environment to the real-world robot arm.

Main Task 1. In clutter problem, selectively grasp the target object and transfer it to the designated location

Main Task 2. In clutter problem, organize the objects in the same arrangement as the given figure

Main Task 3.In clutter problem, build blocks in the same arrangement as the given figure

Artificial intelligence-based algorithm development for sleep analysis and diagnosis of sleep apnea syndromes (2018~2019)

We are developing deep learning algorithm for automated sleep stage scoring and diagnosis sleep-apnea of syndromes.

Automation of Sleep Stage scoring

Using single-channel EEG signal, our deep learning model called IITNet automates the sleep stage scoring by extracting and analyzing the features at sub-epoch level with the state-of-the-art performance.

Non-contact Respiratory monitoring

We are developing a non-contact respiratory measuring algorithm which predicts the respiratory signals from UWB signals in real-time via a convolutional neural network.

Sleep apnea diagnostic system

A sleep management system that automatically detects sleep apnea by integrating the deep learning models for sleep staging and respiratory monitoring.

Development of Assistive Technology in Elderly Care (2018~)

Because of increasing the number of elders, the elderly care labor force becomes a more important issue. Considering this problem, the robot with AI will help this.

Monitoring the taking medicine based on video

Using Deep Visual Tracker, the robot will track the elder. Then, using deep CNN and LSTM, it will classify the user and judge behavior of taking medicine.

Sensing the emergency situation

By database of user’s normal behavior, we can get the network which can have the same output with the input learned before using Deep Auto-encoder. Using this, we can classify between normal behavior and abnormal behavior.

Prediction of the hazard of dementia and stroke

We can judge the diseases(dementia, stroke, etc.) by the elder’s walking motion. Therefore, we have to collect the skeleton data using the Depth Camera, and then develop an Anomaly Judgement Algorithm based on Deep Auto-encoder.

Collection of big data related to elder using sensor

We have to develop the server software which can collect and manage the database in real time. This software can make dataset to be used to Deep learning.

QUICK
MENU
GIST 대표GIST 대표 GIST PortalGIST Portal 융합기술학제학부융합기술학제학부 도서관도서관 증명서 발급증명서 발급