November 13 (Tuesday), 2018
VP, Automotive, Synopsys, USA
Dr. Burkhard Huhnke is the Vice President of Automotive Strategy at Synopsys. He joined Synopsys earlier this year.
Prior to Synopsys, he was SVP of Product Innovation & E-Mobility at VW, based in Silicon Valley. He was responsible for synchronizing VW's innovation activities and alliances to identify new concept ideas, business models and partners in the US and had end-to-end ownership of the electric vehicle platform in North America.
Prior to that, he held several positions both in the US and Germany, including Senior GM, Electronics System Integration and Whole Vehicle Integration.
Dr. Huhnke studied electrical engineering, at the University of Braunschweig. His dissertation about optical distance measurement was awarded with the International Measurement Prize.
Dr. Huhnke serves as Research Fellow the Hult Business School in San Francisco, and is a member of the Board of Advisors at the College of Engineering at University of Tennessee Knoxville and at the College of Engineering and Computer Science at University of Tennessee Chattanooga.
In the automotive world, recalls for electronics affects about five percent of the vehicles on the road. That means 5 out of every 100 vehicles today have a problem with their electronics. If we want to see more autonomous driving vehicles, that number must be improved. There must be more robustness into the development process.
Unlike a blue screen of death that may occur on a desktop or laptop computer, a chip or software failure within a vehicle traveling at highways speeds could result in significant injury or death to the driver, their passengers, or others.
Preventing this possibility requires building in functional safety. Specifically, ISO 26262 sets out definition for various Automotive Safety Integrity (ASIL) levels with Level D (ASIL-D) being the highest automotive software integrity level, and often the hardest to achieve. To guarantee the safety of any vehicle, ASIL-D must be a part of even the smallest pieces of electronics.
Robust design can and should begin at the SOC level by integrating functional safety from automotive industry. In other words, functional safety doesn't have to wait until later -- in the system world, in the ECU world – it can be built into the chip design itself. This can be done with a virtual prototype, a model of the hardware in development, which can allow software development on those SOCs to begin at a very early phase.
Implicit is the continuous testing of these hardware and software designs throughout the automotive development lifecycle. If the chips and the software can be secured, step by step, within the automotive development process, this creates clear milestones that build in safety and security and quality along the way. It also "shifts left" the mitigation of errors that traditionally compound as the vehicle gets closer and closer to production, the traditional error curve begins to flatten. This, then, accelerates the overall time to production.
The keynote talks about tools to optimize the quality, security, and safety throughout the vehicle lifecycle. For example, safety-island IP and dual-core lockstep processors and the new ASIL-D-ready-certified processors come with a self-checking safety monitor as well as hardware safety features, such as error-correcting code and a programmable watchdog timer to help detect system failures and runtime faults.
The virtual prototype technology allows the entire automotive ecosystem to begin software development in parallel with hardware development – accelerating time to production, which means disrupting the current development process and a shift left by co designing hardware / software for automotive electronics is possible.
The transition to the era of artificial intelligence is rapidly going on, and IT industry is making great efforts to apply the artificial intelligence to the virtually most of services and products. This keynote emphasizes the big wave behind, that is, to meet demands for the intelligent user experiences along with the customer-tailored evolving products.
On the one hand, speech recognition becomes ubiquitous driven by Google and Amazon focusing to maintain and widen their platforms. Even more sophisticated image recognition and context understanding are also emerging to provide much convenient user experiences. On the other hand, the concept of the intelligence has a great potential to revolutionize the performance of the systems, that is, to enhance the fundamental characteristics of products. For example, the picture quality of TV can be enhanced with AI, and the energy efficiency of home appliance products can be improved with AI.
LG has been successfully launching various market-leading products such as OLED TV, and home appliances. As a system design/manufacturing company, LG seems to have more advantages for making innovative customer-tailored products with AI utilizing huge amount of accumulated data. Intelligent SoC plays a crucial role for the innovative products. This talk introduces LG intelligent SoC R&D activities, and achievements.
Dong Sam Ha
A smart wireless sensor node (WSN) built on SoC processes sensed data locally and transmits it wirelessly. A smart WSN will be more pervasive in the era of IoT (Internet of Things). A major design issue for WSNs is autonomous power, and energy harvesting is a promising solution. Energy harvested from ambient sources aims to recharge the battery of a WSN or even remove the battery perpetually. Typical energy harvesting sources include solar, thermal, kinetic, and radio frequency. The operating environment of ambient energy sources changes and the energy level of typical ambient energy sources is low. They poss major design issues for energy harvesting circuits, which aim to maximize the energy flowing into storage devices such that rechargeable batteries or supercapacitors. This talk covers design issues of energy harvesting circuits for various energy sources and major design schemes to address the problems. It also covers recent advances and research issues in energy harvesting circuits in the context of smart WSNs built on SoC.