AI and Neural Networks: Essential Elements for Autonomous Driving Applications
Infineon Integrates Synopsys PPU into AURIX MCUs
PPU accelerates AI algorithms RNN, MLP, CNN, and RBF.


AI and neural networks are becoming increasingly important in the development of safe, smart, and environmentally friendly cars.

Infineon Technologies announced a collaboration with Synopsys to support AI-based solutions with next-generation automotive MCUs.

The next-generation Infineon AURIX MCUs are integrated with a high-performance AI accelerator PPU (Parallel Processing Unit) that features Synopsys' DesignWare ARC EV processor IP.

AI and neural networks are essential components for autonomous driving applications such as object classification, target tracking, and path planning. They also optimize various automotive applications, reducing ECU system costs while increasing performance and shortening development times.

For example, it enables optimized engine auto-calibration and reduces the number of sensors by creating accurate mathematical models of the physical reactions occurring in the system. Instead, AI requires much more computing power than standard algorithms.

AURIX already supports specific neural networks, but the PPU's higher performance than current accelerators will enable real-time processing of sensor data that is currently limited to real-time processing.

PPU will accelerate AI algorithms such as Recurrent Neural Network (RNN), Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Radial Basis Function (RBF).

Synopsys' "MetaWare EV Development Toolkit for Safety" is available for the new EV processor. This toolkit enables rapid development of application software that meets automotive safety requirements.

The AURIX tool chain supports model-based design, enabling you to leverage the latest software design techniques and shorten increasingly demanding automotive development schedules.

Additionally, PPU supports CNN, which helps move towards a comprehensive security system. It will enable layered security by supporting intrusion detection and defense systems such as deep packet inspection and system entropy monitoring.

“Safety is paramount in many AI-based applications,” said Joachim Kunkel, Head of Solutions Group at Synopsys. “By combining the processing performance and safety capabilities of our ARC EV processor with the proven architecture of AURIX, we will be able to develop automotive systems that meet the highest functional safety requirements.”

“By developing the PPU together with Synopsys, we are ensuring that Infineon’s next-generation MCUs will offer the safety features, throughput and power efficiency to meet the increasing computational requirements of AI,” said Peter Schäfer, head of the Automotive MCU business line at Infineon. “We are also preparing AURIX for future data-intensive automotive applications such as gateways, domain controllers, engine management, e-mobility and ADAS.”