What challenges will automotive IC and electronic and electrical architecture design face in 2022?

At this stage, the automotive industry is gradually migrating to a new architecture, with higher requirements for security, data transmission speed, and cost control.

The automotive industry faces severe challenges both in the short and long term. In the short term, chip shortages remain a top concern for global automakers. The chip shortage has severely delayed the delivery of new cars, interrupted the repair of existing vehicles, and greatly inhibited the growth of the automotive industry. As upstream wafer fabs continue to expand production capacity, supply chain issues are gradually being resolved, but it will still take some time for the automotive market to truly achieve a balance between supply and demand.

"Supply chain shortages are the most significant issue in the overall automotive market in 2021, and this situation is expected to continue into 2022," said Robert Schweiger, director of automotive solutions at Cadence. "Based on this, companies are beginning to seek alternative solutions to address chip shortages, including developing their own chips and establishing strategic partnerships with foundries."

The partnership between Ford and GlobalFoundries is a prime example. “What this means for 2022 remains to be seen, but it shows that supply chain shortages are having some impact on decision-making, partnerships, and future strategies of automotive companies,” Schweiger said.

The chip shortage has gradually made chip procurement and engineering talent the focus of the industry, leading to longer-term and more fundamental changes in the automotive industry. "OEMs and Tier 1 will increase their internal engineering expertise in software and hardware," said Walter Wottreng, vice president of the automotive division at Synopsys. "Engineers will introduce more flexible electronic and electrical architectures to quickly rewrite software based on supply chain needs and integrate alternative chips. In addition, increasing internal expertise will help OEMs better cope with supply chain pressures. As a result, they will have a better understanding of chip availability and establish a more direct relationship with the semiconductor supply chain."

In fact, automakers and Tier 1 are improving their ability to cope with chip shortages, for example, by developing software to leave room for themselves. "Having learned lessons from the ongoing chip shortage, semiconductor suppliers are becoming more active in the software field to make them more adaptable to the latest silicon designs," said Artur Seidel, vice president of Elektrobit Americas. "They provide chips with dedicated software. This coupling includes functional safety/information safety (safety/security) components, so it is a win-win for customers. For example, Qualcomm acquired Veoneer and Renesas Electronics provides gateway solutions and software packages."

In addition, a series of ongoing transactions also reflect new developments and changes in the entire automotive industry. "By 2022, the automotive industry ecosystem will gradually improve, and semiconductor suppliers, subsystem providers and OEMs will establish more collaborative ecosystem partnerships." Lars Ullrich, vice president of automotive in the Americas at Infineon, said. "We see that demand for automotive semiconductors will continue to be strong in 2022, and the world will promote the digitalization and electrification of future vehicles."

What changes have taken place in the automotive industry?

One of the major changes in the automotive industry is the focus on vehicle safety, especially software safety, due to the increase in highly automated driving (HAD), vehicle-to-everything (V2X) and software-intensive systems, Wottreng said.

Lee Harrison, manager of automotive IC test solutions at Siemens EDA, also pointed out that multi-layer security and product reliability are one of the biggest challenges facing automotive ICs.

“Security is a growing area of ​​concern. There is increasing research and investment to try to address and understand security issues. Additionally, reliability and data acquisition are critical for new advanced ADAS systems that use the latest technology nodes and have no historical reliability data to rely on,” said Harrison. “Automotive applications need to be designed with security in mind to ensure that sensitive data cannot be accessed by external agents. In the past, this was a somewhat specialized requirement, and it was common to implement custom solutions to meet these specific requirements. However, in recent years, with the explosive growth of the semiconductor industry in the automotive and cyber-physical systems sectors, the need around security testing and monitoring has become mainstream. Today, chip manufacturers need to extend security technologies into multiple different levels of SoC development to provide the best coverage in a defense-in-depth solution.”

"Even if everyone says that information security is never the primary business driver, it is always secondary to the primary driver," said Andreas Kuehlmann, CEO of Tortuga Logic. "The primary driver is more like functional safety in automobiles, and in medical devices, it is privacy and medical records. When we talk about regulated industry compliance, information security is always an essential element required for these major business goals. For the automotive industry, the biggest wake-up call was the Jeep Cherokee incident, where two hackers successfully hacked into the car and remotely controlled the engine, steering system, etc., causing the vehicle to lose control."

Kuehlmann pointed out that this is a wake-up call for the entire automotive industry, and also a positive sign. "They realize that automotive safety is not just a functional safety issue. Unlike the general reliability of the car, safety is like a switch. For damaged mechanical parts, it usually means that a small part of them are bad, but most of them are intact, but this ratio changes over time. The automotive industry is actually very good at measuring and monitoring safety risks and recalling cars. If the car exceeds a certain threshold of damaged parts, if it is an airbag or other aspects of the vehicle, the car OEM will conduct a large-scale recall. This is a purely economic decision," he said.

Cybersecurity changes the equation. "If one particular model of car is vulnerable, then all cars on the road are vulnerable, which means you can hack all of them at once," Kuehlmann said. "The problem is new and not just in the automotive space. That's why it's so hard for boards to understand what cybersecurity means. Traditional risk management tools don't apply directly to this. You can do all the prevention you want, but if someone publishes how to hack car XYZ and posts it on the internet, all of a sudden all cars are vulnerable. You can manipulate the car by flipping a switch, and that's pretty basic stuff and something the automotive industry understands."

Automobile Economy 2.0

Standardization management is a key way for the automotive industry to solve these problems. These standards have been proven in the past to effectively control the cost of the entire ecosystem. Taking the next-generation vehicle-level architecture solution as an example, the regional controller pays special attention to automotive industry standardization, and many automakers, including some using advanced process nodes, are actively deploying.

“It will be expensive to develop a regional controller at an advanced node,” said Cadence’s Schweiger. “For the high cost, many companies will amortize the cost by selling it in high volumes. This means that after a company develops a regional controller, it will offer its technology to their competitors and even the industry to increase sales . You can see this in the automotive space – for example, with AUTOSAR – where early adopters considered AUTOSAR as a car operating system, and then they started to come up with their own proposals. Then they tried to bring more automotive companies together to form a consortium and eventually develop a new standard together, an open standard like we have today. Today, hundreds of companies are part of the AUTOSAR organization. The automotive industry is dominated by open standards, not customized solutions. That’s why with regional controllers and regional architectures, we may see again that some companies will face huge challenges and try to come up with their own solutions, which will gradually become open standards for many companies to solve the cost problem.

This also applies to the security economy. “Especially when we talk about cars with OTA updates, V2V communication, V2I communication, cybersecurity is becoming more and more important,” Kuehlmann said. “If we don’t take precautions, if we don’t bake security into the product – not only into the product but into the entire infrastructure – we will face huge security risks. Security is part of the reason why autonomous development has slowed down a bit. The challenges of automotive cybersecurity are extraordinary and different from what we have known in the past.”

But understanding this and planning for it are two different things. Keeping cars safe requires building layers of security into the initial architecture and being flexible enough to adapt to new threats as they become known. For a car that may be on the road for decades, resilience is critical.

This is at the heart of a concept called "defense in depth." According to Siemens' Harrison, defense in depth involves a series of technologies that can be implemented at different levels of the design to build a defense ring that introduces many levels of protection to a device using relatively simple concepts and provides significant defense against malicious attacks.

Some of these concepts are passive, providing a safety lock that prevents access. Other techniques are active, taking evasive action when a threat is detected. This could be as simple as resetting the device if the BiST controller is configured in an unintended mode, or blocking illegal bus tasks based on data collected by the Bus Sentinel analytics monitor.