How To Improve The Reliability Of Self-driving Cars

June 28, 2021

Latest company news about How To Improve The Reliability Of Self-driving Cars

However, as real-world trials take place on today's roads, the range of functions that self-driving cars must support is expanding and becoming rapidly more complex. These automatic systems will constantly improve the performance, power consumption, safety, security and reliability requirements. For automotive Oems, to ensure that self-driving cars comply with safety regulations, they need to design hardware and software in accordance with THE ISO2626 functional safety standard. If developers are ill-prepared, they will need to invest extra money and time to prove that their products meet safety standards, potentially delaying the launch significantly, squeezing profitability and eroding market share.


The core goal of safety and reliability of autonomous vehicles is to prevent personal injury and property damage. When the accident happens and who is responsible for the accident are also legal issues that need to be considered. In such a traffic state, automatic driving is faced with numerous legal problems, and how to determine the attribution of responsibility when an accident occurs is still pending. Therefore, failures must be avoided. This has led auto Oems and auto market suppliers to pay more attention to reliability. Proving that every component in a smart car is safe and reliable is therefore crucial.

latest company news about How To Improve The Reliability Of Self-driving Cars  0


Smarter, more reliable storage

Self-driving cars are equipped with advanced Advanced Driver Assistance System (ADAS). These vehicles have multiple sensors (cameras, lidar, etc.) and controls that allow them to drive autonomously and avoid collisions. These sensors and controls are mission-critical and cannot fail. FIG. 1 shows the schematic diagram of an automatic driving system with level 3, 4 and 5 automatic levels that can drive without monitoring.

latest company news about How To Improve The Reliability Of Self-driving Cars  1


Nonvolatile memory devices play an important role in ADAS systems, providing startup code storage and data logging for important mission-critical events. As these systems become more intelligent, they need to process more data faster and with higher levels of reliability. In addition, even if the ADAS design is otherwise reliable, it can be vulnerable if the memory is not protected (that is, the memory bit is not verified at startup or during device operation).

NOR flash is the ideal memory technology for mission-critical applications because it provides non-volatile storage supported by high reliability and integrated diagnostics. Integrated diagnostics ensures data integrity, detects possible failures, and even corrects errors. In addition, advantages such as just-in-time startup and high performance fast system startup time facilitate immediate access to code, configuration data, and graphic images when the car is powered on.

Today, in order to meet automotive functional safety standards such as ISO26262, the memory device family needs to be designed from scratch. These new generations of memory not only provide greater reliability, but also improve performance, significantly reduce power consumption, and reduce overall cost of ownership.



One of the most effective ways to simplify a system is integration. When a system is composed of many components, each component and its interconnections with other components can be a potential point of failure. For example, integrating the MCU with storage results in faster data and code access, more efficient processing, greater reliability, and lower cost. In addition, development is simplified because components that previously had to be integrated by developers into larger systems can now be managed internally by the MCU.


The benefits of integration now extend to NOR flash. As memory manufacturers begin to integrate memory with processors such as Arm Cortex-M0, complex processing needs to be done to maintain the reliability of high-density, high-speed memory (see Figure 2). The advent of onboard processors could revolutionize the way engineers use flash memory for design by enabling smarter storage. For example, in the past, to extend the life of flash memory, a lot of work has been done on the development of wear equalization software. Now, the loss equalization problem is managed internally by the integrated MCU.


latest company news about How To Improve The Reliability Of Self-driving Cars  2


A new generation of complex SoC using 16nm FinFET technology is not yet capable of embedding flash memory into a chip. So they must rely on smarter and more reliable external NOR flash technology. Not only can onboard processors be used to manage all the security-critical areas of memory storage, but they can also be used to manage the network security area of memory to prevent malicious attacks. When integrated processors are incorporated into flash memory, these units are self-managed by memory devices and can be rapidly configured to meet specific application requirements.


Changing demands

At present, the automotive industry is moving from driver assistance to fully automated development. These systems will require intelligence at all levels to reduce delay and improve efficiency. At the same time, the internal architecture of the automobile is also developing from the main independent discrete system to the interconnected system. Interconnected systems can transmit data between systems in real time and play the role of artificial intelligence and machine learning. In addition, the data collected from the vehicle will be used to implement predictive maintenance so that the vehicle can prompt the driver to maintain the vehicle before a failure occurs. In order to do more sophisticated analysis and complete a new software upgrade from the cloud to the car, you also need to send data to the cloud.


Intelligent flash storage is at the heart of these systems because critical code and data stored in these non-volatile memories still need to be reliable and last for more than 20 years without failure in extreme environments. By adding onboard processors, these memories can now provide a higher level of functionality and reliability, while offloading memory management tasks such as loss balancing, enhancing system security with cryptographic protection, and performing security-critical diagnostics.

Autonomous driving is a rapidly growing industry, and new safety features and safety assurance features will be developed and standardized at the same speed. Oems need a flexible architecture that ADAPTS to these standards in a timely manner and introduces advanced features that enhance long-term reliability. For example, when memory can predict a specific type of failure, it can begin to prioritize.


To help automotive Oems build compliance systems, memory manufacturers need to provide ISO 26262 compliant security documentation, including detailed safety analysis reports such as safety manuals, failure mode impact and Diagnostic Analysis (FMEDA), dependent Failure Analysis (DFA), and context-independent security elements (SEooC). In addition, memory manufacturers need to actively develop and comply with these standards to ensure that their components continue to comply with regulatory requirements.


Memory devices such as Cypress's Semper NOR flash are designed to meet the challenges of a new generation of vehicles and industrial systems and meet a wide range of quality, reliability and safety standards.