DocumentCode :
3722766
Title :
A Real-Time Streaming System for a Multifunctional Vehicle Information Display Based on TMO
Author :
Dongmin Kim;Kimoon Hong;Byeongyong Jung;Jungguk Kim
Author_Institution :
Dept. of Comput. Sci. &
fYear :
2015
Firstpage :
210
Lastpage :
215
Abstract :
In recent vehicles, multifunctional displaying of various data from Electronic Control Units (ECUs) and multimedia devices is required for monitoring or for safe driving. Control Area Network (CAN) is generally used for transmitting status data such as diagnostic information of a vehicle. For automotive audio/video data, either Media Oriented Systems Transport (MOST) or 1Gbps synchronous Ethernet is used. And also, several types of out-vehicle and in vehicle cameras are attached to be used for infotainment, logging and intelligent safe driving. In this paper, a real-time streaming system for multifunctional displaying of two types of vehicle information is presented. The first data type is the vehicle status information from an ECU via CAN. The second one is the Motion JPEG (MJPEG) video data from an Ethernet camera device. After gathering CAN and MJPEG frames, the streaming system combines them into messages in a new frame format and streams them to a multifunction display device in real-time. In order to provide timeliness of streaming of two different types of frames with different bandwidths, several real-time data acquisition tasks and a real-time streaming task are necessary. Of course, a suitable deadline-driven real-time scheduling for each task is also required according to the timing characteristic of the task. For the modeling of the timeliness-guaranteed system, the Time-triggered Message triggered Object (TMO), a well-known real-time distributed object model, was used in modeling the system. As an execution engine for TMOs, a middleware for real-time distributed computing on Linux, named Linux TMO System (LTMOS) v2.0, was used. Considering that 1Gbps Synchronous Ethernet is emerging as a next-generation backbone for automotive networks, the developed system is expected to be used widely in the developments of multifunction displays for vehicles in the near future.
Keywords :
"Real-time systems","Vehicles","Streaming media","Object oriented modeling","Linux","Computational modeling","Cameras"
Publisher :
ieee
Conference_Titel :
Knowledge and Systems Engineering (KSE), 2015 Seventh International Conference on
Type :
conf
DOI :
10.1109/KSE.2015.40
Filename :
7371784
Link To Document :
بازگشت