2025 SMSI Bannerklein

P13 - AI-supported PDN Design for PCBs in Automotive Applications

Event
iCCC2024 - iCampµs Cottbus Conference
2024-05-14 - 2024-05-16
Cottbus
Band
Poster
Chapter
Mobilität
Author(s)
T. Bartels, D. Choy, B. Schröder, B. Stube, A. Pucic - Technische Universität Berlin, Berlin
Pages
157 - 160
DOI
10.5162/iCCC2024/P13
ISBN
978-3-910600-00-3
Price
free

Abstract

Due to market trends such as electronic vehicles and autonomous driving, the technological complexity of electronics for automotive applications is increasing. The rising development requirements motivate the usage of AI modules in PCB design to obtain functionally secure and reliable layouts faster. This contribution presents a prototype PCB layout tool with AI-supported features for the design of power delivery (PDN). It supports PCB developers with real-time feedback
for the choice of types and positions of decoupling capacitors using Convolutional Neural Networks (CNN), suggests PDN-designs using Reinforcement Learning (RL) agents, and integrates into workflows with proprietary design tools that are common in industry-use such as ZUKEN eCADSTAR or Altium Designer.

Download