Horizon Robotics Journey™
The new AI Inference processor
Purpose built Edge-AI processors for Autonomous Machines
Scalable processors crafted for AI performance, flexibility, cost and energy efficiency
Efficient Deep Learning
Revolutionary Bernoulli2 dual-core Brain Processing Unit with 5 TOPS to accelerate AI workloads
Powerful ISP
With HDR support, Journey ISP delivers high-quality images in difficult conditions at low latency
System-on-Chip
Balanced architecture with quad ARM Cortex-A53, deterministic ethernet, USB 3.0 and 4K60 H265 codec in a small 16nm die
AI Optimization
Harmonizes throughput, low latency, accuracy, energy efficiency and cost effectiveness
Vision input at 30fps
DL INT8 peak compute
Typical BPU utilization
Typical low power in ADAS
Designed for AI compute
Perception workloads
Designed for the optimum runtime execution of Vision and LiDAR perception tasks
Benchmarks
Outperforms competition in benchmarks of modern neural networks such as Mobilenetv2 and EfficientNet Bx in latency and throughput
Open AI Toolchain
Effectively enables your deep learning model to be optimized for the BPU CNN engine
Automotive grade
Complies with automotive quality, reliability and safety standards, including AEC-Q100 grade 2
True AI performance measured in accuracy, speed and efficiency
Solutions Served
Journey AI processors are scalable from ADAS to Robotaxi applications as well as in-cabin AI systems. Journey 2 has won the coveted Product-of-the-Year award in the category of Automotive Solutions in June 2020
