On January 26, Wu Jihan, co-founder of Bitmain, officially announced his resignation as CEO and chairman of Bitmain through an encrypted letter. The rhythm BlockBeats unraveled the content of the letter through the password posted by Wu Jihan on Twitter. Jihan Wu explained the final solution to the struggle with Ketuan Zhan in English. At the same time, he also firmly believed that Bitmain would s...
“The number of people buying new energy vehicles (EVs) is exploding. It is estimated that the number of new energy vehicles on the road will reach 10 million by 2030 and 36 million by 2040. The auto industry is at an inflection point as governments set carbon emission targets and as more consumers realize the benefits of new energy vehicles in terms of cost of ownership and environmental protection. This change comes at the same time as advanced driver assistance systems (ADAS) and autonomous driving features are coming.
The number of people buying new energy vehicles (EVs) is exploding. It is estimated that the number of new energy vehicles on the road will reach 10 million by 2030 and 36 million by 2040. The auto industry is at an inflection point as governments set carbon emission targets and as more consumers realize the benefits of new energy vehicles in terms of cost of ownership and environmental protection. This change comes at the same time as advanced driver assistance systems (ADAS) and autonomous driving features are coming.
With the increasing popularity of LV3 and LV4 autonomous driving technology, drivers can take their attention away from the road and not worry too much about driving. There is a growing demand for sensors that will be integrated into vehicles to replace the driver’s eyes. More and more cameras, external LiDAR (LIDAR) and even internal sensors to monitor driver safety and attention will undoubtedly be a good thing for the future of driving, but one thing is very challenging: the consumption of electricity.
Every watt of power consumed by a sensor affects the vehicle’s range, so how can automakers implement the characteristics of the car of the future without affecting the vehicle’s range?
Considering the design of new energy vehicles
An easy way to do this is to improve the efficiency of the system itself. By developing low-power sensors that can collect data, we can immediately save power consumption, but all collected data needs to be transmitted to a processor for processing. The more data, the greater the power consumption of the chip processing. Advanced concept cars for autonomous driving and advanced ADAS systems rely on PC-level water-cooled GPU solutions to handle these intensive and complex computations. But when the capacity of the battery is limited, the GPU with 350W power consumption is stretched.
One of the best ways to solve this problem is to design from the ground up, thinking about the car as a platform. Rather than using multiple modular designs, higher-density chips can be used to enable more efficient computing while reducing the number of chips in the car. With a custom design, functional safety can be designed from the ground up. For example, the recent Imagination XS series GPUs designed for automotive are ISO 26262 compliant. This means that a single GPU can have immediate failure detection internally. Automakers need 50 percent less chips compared to other solutions.
From macro to micro, magnify small gains
In our main ADAS GPU products, we have custom-designed, high-density, functionally safe chips, but how else can we reduce power to allow drivers to drive more miles without turning everything off? Hundreds of Electronic control units (ECUs) are integrated in any modern car, spread across the various modules of the car. These units can control and monitor high-end ADAS functions and even control the car’s power mirrors. Each microcontroller needs to control the power budget of its own function in the entire car system, and one or two microcontrollers may have an impact, but the overall power number is very large. Reducing the number of ECUs in a car is also an aspect of functional safety considerations. Each core needs to manage a separate application or function to ensure no information crossover and functional safety of the entire system. IP vendors are studying how to challenge this in a safety-focused way. traditional practice.
One way to reduce power consumption is to turn the functions controlled by different ECUs into a multi-core ECU processing unit. By combining multiple functions, whether it is positioning or other scenarios, we can reduce the need for independent ECUs. The resulting power consumption. Hardware virtualization is an option that allows the GPU to perform multiple workloads, such as the dashboard, rear-view camera, and infotainment control center all have a single chip to handle it. By isolating these functions on a single GPU, multiple operating systems and functions can run independently at the same time, and by meeting functional safety standards such as ISO 26262 to perform self-checks on each core’s failures, we can easily see that the number of ECUs has greatly increased. reduce.
keep moving forward
New energy vehicles are not widely used, and the industry has not yet formed a safer autonomous driving standard. From ADAS to door locks, every watt of power is critical. The key to integrating new energy vehicles with autonomous driving technology is IP suppliers and chip manufacturers, who will continue to develop and support various functional features for automobiles. Provide solutions for achieving low power consumption.