BROWSE
Detail
Conference Paper
A Heterogeneous and High Performance Driving Computing Hardware Platform for Autonomous Vehicles
Cited - time in 
Share 
Share
- Authors
- Jung-Hee Suk, Chun-Gi Lyuh, Young-Deuk Jeon, Min-Hyung Cho, SungHoon Kim
- Issue Date
- 2018-05
- Citation
- IEMEK Symposium on Embedded Technology (ISET) 2018, pp.1-2
- Publisher
- 대한임베디드공학회
- Language
- English
- Type
- Conference Paper
- Abstract
- This paper describes a high performance computing hardware platform composed of CPU, GPU, and FPGA based heterogeneous computing structure capable of real time processing of autonomous driving algorithm such as object recognition, pass planning, and vehicle control by receiving data from camera, radar, and lidar. The platform supports eight channels of FHD cameras, five channels of radar and ladder, four channels of CAN for vehicle control, and an FPGA-based preprocessor and convolutional accelerator for deep-learning algorithms.
- KSP Keywords
- Autonomous vehicle, Hardware platform, Heterogeneous computing, High Performance Computing, Object Recognition, Real-Time processing, Vehicle control, autonomous driving, deep learning(DL), learning algorithms