Efficient System-on-Chip (SOC) for Automated Driving with High Safety

Yutaka Yamada,Katsuyuki Kimura

DOI: https://doi.org/10.1007/978-3-030-18338-7_29

2020-01-01

Abstract:The requirements of SOCs for autonomousAutonomous driving systems are described. Huge data from several sensors such as a high-dynamic-range camera, RADARRADAR, and LiDARLIDAR have to be processed efficiently to detect, recognize and track objects such as pedestrians and vehicles. AcceleratorsAccelerator including a deep neural networkNeural network are suitable for energy- and area- efficiency. Complex control functions such as path planning, vehicle control and error handling are important to keep the vehicle stable. Multi-coreMulti-core processors and DSPsDSP can perform these function and have the capability for update features. The energy- and area- efficiencies must be improved continuously as well as operating speeds to achieve more safetySafety. The safetySafety features detect faults and prevent fault propagation, and have to meet standards like ISO 26262. The securitySecurity features are for black-box protection and prevent external attacks. The forecast is concerned with technology progress, the evolution of standards, with multi-sensor inputs and with safe, networked autonomousAutonomous mobility.