Standalone Around View Monitor-based Re-plannable Automated Parking System
(September, 2017 ~ current)
-
Project objective: Development of an automated parking system using only around view monitor in complex parking situations
-
Role: Project manager and algorithm developer
-
Research theme: Deep learning-based semantic segmentation and parking space recognition
-
Neither range sensors and 3D reconstruction algorithms are required
-
Robust slot marking extraction and obstacle detection from the unified framework
-
Ability to recognize parking spaces and occupancies simultaneously
-
System architecture of automated parking system
Original video
Semantic segmentation
Demonstration of the proposed automated parking
Development of Traffic Jam Assist System for EV based on Affordable Sensors
(September, 2016 ~ August, 2019)
-
Project objective: Development of a traffic jam pilot function with autonomous capabilities using low-cost automotive components operating in France and Korea
-
Role: Project manager and system engineer
-
Founded by: Ministry of Trade, Industry and Energy
Development of Highly Automated Driving System Based on Automotive Sensors
(September, 2013 ~ February, 2017)
-
Project objective: Development of highly automated vehicle for highway and urban driving
-
Role: Project manager and algorithm developer of traffic light recognition
-
Founded by: Hyundai MOBIS
-
Research theme: Traffic light recognition system using a high definition map
-
An effective architecture that integrates a vision system with an accurate positioning system and an extended digital map.
-
Publication of Journal: Jang, Chulhoon et. al. ."Traffic light recognition exploiting map and localization at every stage" Expert Systems with Applications 88 (2017): 290–304.
-
Patent: Method and device for detecting traffic light at slanted road (10-2016-0102768)
-
Traffic Light Recognition for three bulbs under the bridge
Traffic Light Recognition for four bulbs on the uneven road
Development of 3D-LIDAR Sensor System for Distance Resolution of the 10cm level
(January, 2012 – May, 2014)
-
Project objective: Sensor development of 3D flash LIDAR to be able to provide object information including pedestrians, vehicles, bicycles, and static obstacles
-
Role: Project manager and algorithm developer of object detection and tracking
-
Founded by: Korea Ministry of Trade, Industry & Energy, Hyundai MOBIS
-
Research theme 1: Object detection and tracking for various types of objects based on the 3D flash LIDAR sensor
-
3D reconstruction from a depth image
-
Measurement generation using K-means clustering and Kalman filter-based object tracking
-
-
Research theme 2: Design factor optimization of 3D flash LIDAR based on geometrical model
-
A novel approach for design factor optimization of the 3D flash LIDAR based on a geometrical model by using structural similarity between the 3D flash LIDAR and 2D digital camera
-
Publication of Journal: Jang, Chulhoon et. al. “Design factor optimization of 3D flash lidar sensor based on geometrical model for automated vehicle and advanced driver assistance system applications” International Journal of Automotive Technology 18, no. 1 (2017): 147–156.
-
Patent: Sensor array optimization for 3D Flash LIDAR (10-2014-0136770)
-
