AutoCAS

Nowadays, autonomous mobility circuits and systems (AutoCAS) have been attracting much attention because of the rapid development of electronic systems-based autonomous vehicles. They are not only used in controlling the engine but also have emerged in the areas of safety, driving assistance, and communications. Autonomous vehicle needs highly reliable circuits and systems for artificial intelligence (AI), sensing, signal processing, and V2X, etc.
Additionally, future electronic systems are moving towards making vehicles that are possible to configure the behaviors by using software easily. Circuits and systems design for autonomous mobility is an area of high investment for semiconductor companies targeting high revenue and growth potential. Our CAS society needs to increase opportunities for our members to communicate/collaborate with researchers/engineers in emerging autonomous mobility CAS directions and contribute to technological innovation and excellence.

Therefore, this workshop aims at announcing the contributions in solving the CAS-based problems for the autonomous mobility components, such as memory, sensors, ECU and deep neural network (DNN), processor, etc. Through this workshop, the researchers are able to obtain the benefit from these innovative techniques. By learning about the advantages and disadvantages of state-of-the-art methods. They can further innovate and enhance their circuits and systems for highly demanding vehicles in five to ten years. The workshop will consist of one plenary track with 6 invited speakers from industry and academia.

 
AutoCAS workshop will be accessible free of charge for IEEE CASS members.
Note. if you are a registered AICAS conference attendee, you will be able to access the virtual option through the conference platform and you do not need to register. All others should register here.

	Compact RF CMOS Radios for Emerging 5G/6G V2X Wireless June 15th, 13:20-14:00(GMT+9) Prof. Mohammad Ismail Wayne State University, USA Biography Abstract Biography Mohammed Ismail , Ph.D., is professor and Chair of the Electrical and Computer Engineering Department at Wayne State University, Detroit and the Founding Director of the WINCAS center of Excellences. He spent over 30 years in academia and industry, having worked in the United States, Canada, Sweden, Egypt, and most recently at Khalifa University in the United Arab Emirates. Ismail is the founding director of Ohio State University’s Analog VLSI Lab, one of the foremost research entities in the field of analog and RF integrated circuits. He was a research chair at the Swedish Royal Institute of Technology (KTH) and created the RaMSiS (Radio and Mixed Signal Integrated Systems) Research Group. He has served the Institute of Electrical and Electronics Engineers in many editorial and administrative capacities, and has co-founded several startup companies in semiconductor and IC design services. He received the 2018 UNESCO medal for his contributions to the development of Nanotechnology. He is a fellow of IEEE. Abstract With the emerging 5G/6G wireless standards, it has become more important than ever to achieve the highest level of CMOS integration for wireless systems on chip. This requires the development of programmable low power multi-band CMOS radios achieving maximum reduction of chip area. This presentation will discuss radio architectures with a focus on direct conversion zero-IF CMOS radios. A compact design sharing the use of transmit and receive basebands for time division duplex (TDD) receivers is introduced. Modeling of the DC offset problem in direct conversion receivers is presented and mixed signal solutions to mitigate or remove such DC offset is introduced. The presentation concludes with CMOS design and implementation of some key CMOS baseband blocks including a time interleaved ADC for 5G V2X wireless communications in 22nanometer FDSOI.
	Computing Platforms for Next Generation Vehicle June 15th, 14:00-14:40(GMT+9) Hwang Gunn Infineon Technologies Korea, Korea Biography Abstract Biography Hwang Gunn has been working for Technical Engineering Center in Infineon Technologies Korea. He is currently leading the technical support team for vehicle automation and chassis applications, focusing on ADAS, Braking and Steering systems based on microcontroller products. Formerly, he had worked for Hyundai Motor Group as an R&D engineer. Abstract In aligned with the major automotive market trends, i.e. ADAS/AD, connectivity and eco-friendly systems, in-vehicle E/E architectures are transforming now and the new computing platforms are required to enable this latest market trends. In this session, the E/E architecture transformations from the application specific domain controllers to a zonal architecture with powerful central computing platforms will be explained. And a heterogeneous computing cluster architecture addressing different computational workloads will be introduced and fail operational system required for autonomous driving will be also addressed.
	Embedded Artificial Intelligence Technology for ADAS/ADS Applications June 15th, 14:40-15:20(GMT+9) Prof. Jiun-In Guo National Yang Ming Chiao Tung University, Taiwan Biography Abstract Biography Prof. Jiun-In Guo received the B.S. and Ph.D. degrees in Electronics Engineering from National Chiao Tung University, Hsinchu, Taiwan, in 1989 and 1993, respectively. He is currently a Distinguished Professor of the Institute of Electronics, the Associated Dean of Electrical and Computer Engineering, and the Director of Embedded Artificial Intelligence Research Center, National Yang Ming Chiao Tung University, Hsinchu, Taiwan. Prof. Guo has been the Director of Institute of Electronics, National Chiao-Tung University during 2013-2015, and served as the Director of SoC research center of NCTU during 2016-2019. His research interests include images, multimedia, and digital signal processing, VLSI algorithm/architecture design, digital SIP design, SOC design, and intelligent vision processing applications including ADAS/Self-driving vehicles. In the past 10 years, Prof. Guo dedicated on researching of the embedded AI technology for ADAS/ADS applications, developed over 30 technologies for ADAS and conducted over 150 industrial collaboration projects. Prof. Guo received the outstanding electrical engineering professor award from the Chinese Institute of Electrical Engineering in 2010, the outstanding engineering professor award from the Chinese Institute of Engineers in 2014, the outstanding research award of Minister of Science (MOST) in 2017, as well as two outstanding technology transferring awards of MOST in both 2018 and 2020 with the topic of AI-based ADAS systems. Prof. Guo is the author of 252 technical papers on the research areas. Abstract This talk regards about how to design deep learning model for embedded AI SoC computing platform dedicated to ADAS/ADS applications, covering the data acquisition and fast annotation, light weight model design and simplification, model quantization for being realized in an AI SoC, and a hybrid fixed point deep learning accelerator (DLA). Some ADAS/ADS design examples will be used for illustration to demonstrate the effectiveness of the mentioned embedded AI technology.
	CMOS Single-photon Avalanche Diodes for Vehicle LiDAR June 15th, 15:30-16:10(GMT+9) Prof. Sheng-Di Lin National Yang Ming Chiao Tung University, Taiwan Biography Abstract Biography Sheng-Di Lin received his B.S. and M. S. degrees in physics at National Taiwan University in 1992 and 1994, respectively, and his Ph.D. degree in electronics engineering at National Chiao Tung University in 2002. In 2002-2005, he worked as a post-doctoral research associate in Cavendish laboratory of Cambridge University in United Kingdom. Joining Department of Electronics Engineering at National Chiao Tung University in 2005, his current research focuses on atomic-scale metallic films grown by molecular beam epitaxy and single-photon avalanche diodes in CMOS technology. Abstract High-performance LiDAR is a key module to enable autonomous driving. In this talk, I shall address the issues hinder LiDAR from current vehicles and how CMOS SPAD would help to address these problems including high laser power, short ranging distance, production cost, system complexity, and power consumption. Starting from a brief introduction to the single-photon avalanche diodes (SPADs) in CMOS technology, I shall present out works on device, array, circuits, and LiDAR system. A few topics regarding photon-counting LiDAR, such as minimum integration time and time-bin width of the time-to-digital converter, will also be discussed. We believe that this ultimate optical sensor could make cost of LiDAR modules effective and available to each car in the near future.
	Holistic Approaches to Memory Solutions for the Autonomous Driving Era June 15th, 16:10-16:50(GMT+9) Dr. Daeyong Shim SK Hynix, Korea Biography Abstract Biography Education: BS (Electronics Engineering at KAIST) 1988 – 1992. MS (Electronics Engineering at Seoul National University) 1993 – 1995. Ph.D (Electronics Engineering at Seoul National University) 2009 – 2013. Work experience: Professor of SKHU( SK Hynix University)(2022~) VP, Head of Automotive Business Unit in SK Hynix (2019 – 2021) VP, Head of HBM Business Unit in SK Hynix. (2016 – 2018) VP, Head of Design Verification & Analysis Group in SK Hynix (2015~2016) Awards & Honors: Award for leader of future top 100 technologies, National Academy Engineering of Korea (2017) National Productivity Award (mobile DRAM production), Korea Productivity Center (2014) Abstract As DNNs improving state-of-the-art accuracy on many artificial intelligence (AI) applications such as computer vision processing for autonomous driving, the data processing bandwidth and power consumption between neural network accelerator and the off-chip memory are big challenge to enhance the compute performance metric TOPs/watt. To overcome the limited compute and energy resources in automobile environment, inferencing by PIM (Processing in Memory) or AiM (Accelerator in Memory) which deployed MAC(Multiply and Accumulation) units and activation function inside DRAM is one of the key solution by using multi bank parallelism and memory cell architecture. When memory technology equipped with analog logic inside mature in the near future, ultra-low power analog accelerator based neuromorphic computing architecture will lead the future autonomous driving solution.
	Transforming the Transportation Industry through AI June 15th, 16:50-17:30(GMT+9) Kyungwon Byun Sr. Solution Architect, NVIDIA, KOREA Biography Abstract Biography As Sr. SAE at NVIDIA, Kyungwon Byun lead NVIDIA Auto/Embedded technology support division, especially focusing on the application of AI technologies to autonomous driving and Robot fields. From large-scale AI training to optimization for embedded, he is in charge of AI end-to-end solutions. Formerly, he had worked LG Electronics for as a senior research engineer. Abstract The transportation industry is going through a major transition with vehicles defined not by horsepower but by computing capability. Driving this revolution is AI and the immense possibilities that it brings by powering autonomous vehicles (AV) that will shape the way people and goods are moved in the days to come. AV development process will keep improving to make AI-powered vehicles smarter and smarter over time as software is trained for new tasks, enhanced, tested, and validated, then updated over-the-air. Under the theme of Transforming the transportation Industry through AI, I would like to present how innovation occurs in transportation through AI.