Scalable software for ADAS and Automated Driving
The aiDrive software stack is used by our partners to accelerate the realization of their Automated Driving goals. Due to its modular architecture, aiDrive can be deployed as the core of any automated driving solution, as part of ADAS solutions, or integrated by module into best-in-class custom systems. aiDrive offers robust perception technologies for mono, stereo and fisheye camera setups and enables the fusion of sensor data across cameras, radars, LiDAR-s and other common sensors in automated driving. aiDrive is also hardware, middleware and OS agnostic, supporting easy integration into existing concepts and solutions.
Validation suite from concept to production
Simulation has long been a staple of aviation, which boasts much higher safety levels than road transport. While the automotive industry also relies on simulation, automated driving presents a new challenge. The importance of simulation is illustrated by Euro NCAP announcing it will increasingly rely on simulation for future safety tests. As an end-to-end CI/CD simulation pipeline, aiSim is a purpose-built solution to address this challenge.
CNN Acceleration for Automotive AI
The aiWare NPU (Neural Processing Unit) is developed by engineers working side-by-side with our automated driving teams to create a unique solution targeting high performance L2-L4 automotive-grade real-time AI inference for AD/ADAS. A highly autonomous engine placing minimum demands on the host CPU, aiWare can be either integrated within an SoC or as the primary computation unit of a companion scalable NPU accelerator chip. Core performance, on-chip memory, floorplan, external memory bandwidth and other parameters are all configurable to optimize performance for a wide range of customer requirements. The latest aiWare4 IP core is a 4th generation design, the result of more than 75 engineer-years development in close collaboration with our aiDrive software and NN algorithm R&D teams.