How a LDM-Service Platform Makes Automated Driving Better?

The day has come. Today, we can easily see people driving automated car on the road; not just in the Sci-Fi or 007 movies. The Korea Institute of Civil Engineering and Building Technology (KICT, President Kim, Byung-Suk) has developed a platform that make an automated driving vehicle overcome risks through dynamic information from infrastructure.

In the case of automated driving in urban areas, surrounding cognition failure can occur when a pedestrian is shaded by a bus or a roadside tree. That is a potential risk factor of automated driving, which becomes an obstacle to the realization of fully automated driving. 

A research team in KICT, led by Dr. Hyoungsoo Kim, has conducted a study to develop an I2V LDM (Local Dynamic Map) service platform that recognizes and shares potential risk factors in the shaded area that automated vehicles cannot do. Following study was funded by Korea Agency for Infrastructure Technology Advancement (KAIA) and the host organization of the study was Korea Transportation Safety Authority (TS). 

The platform recognizes and tracks vehicles, motorcycles and pedestrians at an intersection and provides location information messages for automated vehicles in real time. The LDM platform was designed as six types of modules. The roadside detecting module produces information from detectors and traffic signals installed at each intersection. The roadside data processing module transmits the information generated at the intersection to the automated vehicle and the center in the form of messages. In the center (processing module and prediction module), LDM database management and prediction information are generated, and then transmitted to the automated vehicle. The vehicle interface module informatizes the message received from the roadside data processing module into information and transmits it to the vehicle with monitoring in the evaluation module.

Demonstration of the designed platform was conducted. Detectors and traffic signal transmission boards were installed at urban intersections, and SPaT, MAP, RSA, and TIM of SAE (Society of Automotive Engineers) J2735 messages were provided from roadside data processing devices. In the scenario, the LDM platform recognizes a jaywalking boy who is shaded by the parked vehicle, and delivers RSA messages containing a pedestrian location information to the automated vehicle to prevent an accident. The RSA message was successfully transmitted to the vehicle every 10 Hz (0.1 sec.). Considering that the 50 km/hr vehicle moves 1.4 m/0.1 sec., the performance of this platform is believed to be close to the human driver’s cognitive response time. 

Dr. Kim’s research team noted that the proposed platform is designed to share dynamic information between infrastructure and vehicles in real time. Dr. Hyoungsoo Kim said that “KICT is conducting an infrastructure guidance study that induces safe driving by guiding vehicles in conflict sections as a follow-up to this study.”

###

The Korea Institute of Civil Engineering and Building Technology (KICT) is a government sponsored research institute established to contribute to the development of Korea’s construction industry and national economic growth by developing source and practical technology in the fields of construction and national land management.