Electronic Products & Technology

Merging power and data ecosystems enable data centres on wheels

By Denis Solomon, Tektronix transportation market segment leader   

Electronics Test & Measurement Engineering automobiles data date-centre devices edge Editor Pick engineering EVs measurement power test transporation

Integrating these two previously distinct engineering disciplines will require new tools and techniques

The merger of the power and computing segments of the electronics industry is an incipient trend, but it is irreversible, and it has the potential to be one of the most consequential technology shifts shaping our near future.

The drive trains of hybrid and fully-electric vehicles are, of course, characterized by power electronics – some of the most sophisticated power electronics systems devised. These systems work in concert with suites of digital electronics in the service of a wide range of vehicular functions, from driver-assist features to controlling comfort systems in passenger cabins.

Paragons of power electronics

Modern vehicles – paragons of power electronics – are some of the most computationally rich edge devices being made. Power electronics and data center electronics coming together in modern vehicles isn’t just a combination; it’s an integration. That observation is the heart, though not the whole, of the new trend we see.

Integrating these two previously distinct engineering disciplines will require new tools and techniques, and that was the impetus behind the recent acquisition of Elektro-Automatik (EA) by Tektronix, a Fortive company. Tektronix is a leading provider in test and measurement (T&M) solutions, and EA is a leading supplier of high-power electronic test solutions for energy storage, mobility, hydrogen, and renewable energy applications. Together, Tektronix and EA will develop some of the tools necessary to build the most effective, efficient new vehicles.

Source: Tektronix

This is where the electronics industry is heading. At CES 2024, Intel CEO Pat Gelsinger noted that the BOM (bill of materials) of every automobile represented by electronics is 4- or 5 percent today, will be 12 percent by 2025, and will be 20 percent by 2030.

“Everything about the car – as we think about this digital transformation – it needs more compute, it needs faster memory, we have to become more power-performance efficient, and every aspect of it is becoming more silicon-centric,” Gelsinger said. “This transition – as you quote my mentor, Andy Grove – ‘it’s in a period of disruption.’”

Scaling traffic

Today’s vehicles already incorporate a rich complement of sensors, and successive model years will be equipped with more. These sensors produce boggling volumes of data, data that is most valuable when processed in real time. That takes considerable computation power.

Even if it is a stretch to describe any single vehicle as a data center on wheels, what happens when you aggregate all the computational power of every vehicle on the road?

The amount of processing power is staggering, easily exceeding some of the largest data centers. The amount of real-time data that is generated, meanwhile, is unimaginably vast.

The power of data

What is important for this new trend, is that as the amount of data increases, its potential usefulness expands exponentially. Advanced driver assist systems (ADAS) are already helping to keep motorists and pedestrians safer than ever before. But the good that any isolated vehicle’s ADAS systems can do is limited to that vehicle and its viewable surroundings.

What if every vehicle could know what every other vehicle knows? The enhanced safety inherent in isolated ADAS systems is instantly multiplied. Say, for example, any single vehicle detects a pedestrian still in a crosswalk after a light changes, or if the Department of Transportation has just barred access to a street, or if poorly secured cargo from a delivery truck just spilled in the two left lanes. That vehicle would proceed with caution.

Source: Tektronix

That information could be just as useful to other vehicles not in sight however. That could include vehicles about to approach the otherwise unexpected conditions from around a corner, for example, or those heading for the area but still hundreds of yards or even miles away. Those vehicles might, for example, advise their drivers to take alternate routes.

Vehicle-to-everything

This type of activity is illustrative of vehicle to vehicle communications and of V2V’s conceptual cousins (vehicle to grid, vehicle to home, and the broadly inclusive vehicle to everything – V2G, V2H and V2X, respectively).

Automotive manufacturers have already begun designing vehicles with the ability to talk to each other. The expected benefit is sharply reducing collisions with other vehicles (cars, trucks, scooters, bicycles, etc.), pedestrians, and stationary objects.

Some limited level of traffic management should be possible with V2V, with vehicles coordinating among themselves, but what about V2X (vehicle to everything)? What if vehicles can communicate with other systems – notably with municipal systems and other infrastructure?

If municipalities actively tap into real-time V2X road data, it would be possible to perform holistic, city-wide traffic management. Traffic management can become that much more efficient when traffic lights and other road systems are included.

Smart cities

And traffic doesn’t just mean vehicles anymore – if vehicles are detecting pedestrians anyway, it means foot traffic could be tracked and potentially managed too. Imagine being able to dynamically coordinate both vehicle and foot traffic before and after stadium events, or at dinnertime near a local restaurant row.

City resources – traffic control, lighting, emergency infrastructure – can be allocated, moved, activated or deactivated based on need, with the assurance that it’s all being done in response to actual ground conditions at that very moment. Imagine being able to coordinate with commercial delivery vehicles when necessary. It should be possible to use real-time road data to make all emergency response operations (police, fire, medical, disaster, etc.) faster, safer, and more effective.

Data ideally should flow both ways. Data from municipal systems – traffic management, lighting, safety surveillance cameras – can all be used by vehicles to operate more efficiently. Urban planning should become better informed with an ongoing source of rich, real-time traffic and environmental data.

The new trend

This vast new trend we see is in fact a gestalt of these and other trends:

  • Vehicles becoming rich with electronics (including those with internal combustion engines [ICE])
  • E-mobility, including electric vehicles (EV)
  • Vehicle to everything (V2X) communications
  • Smart cities and the internet of things (IoT)

These are familiar trends, true, but they are all being worked on mostly separately. Together, however, they complement and amplify each other in very real ways. They have synergy, in the old, dictionary-definition sense of the word.

V2X is an idea that’s been kicking around for some time, but it has never been widely implemented. There are several reasons for this, but an important factor is that V2X got tangled up with the fitful progress of 5G network build-outs and technology development.

V2X was one of the most important use cases proposed for 5G when it was first being formulated roughly 10 years ago. The build-out of 5G by communications service providers started out slowly, however, and many automotive OEMS began exploring alternative V2X communications options.

Source: Tektronix

Now, many years later, 5G network coverage is sufficient to support widespread V2X in most places; furthermore, some of the newer 5G technology revisions have made 5G better suited for V2X. A growing number of major car manufacturers have begun designing 5G connectivity into their vehicles.

V2G becomes integral to this trend, as EVs will not only draw power from the grid but also return it. During peak demand, data centers and other operations that represent heavy draws on the grid can pull energy from networks of connected vehicles. This will help balance the grid and ideally reduce dependency on carbon-emitting energy sources.

The smart city is another enticing concept that has been slow to develop for the simple reason that few cities have the financial wherewithal to implement and maintain wide-scale new technology projects.

Now, modern vehicles automatically make their environs smart, given their growing capabilities, expanding number, and computational power – with little or no municipal investment required. Cities should be able to tap into the real-time V2X data to become even smarter. Urban areas will become more responsive to citizen needs, optimizing resources, and reducing environmental footprints through intelligent data and energy management.

Power + digital electronics

All of this hinges on the marriage of digital electronics, sensors, processing power, and memory with power electronics in the form of EVs. The fusion of these technologies will be underpinned by artificial intelligence (AI) and machine learning (ML) algorithms. AI/ML will optimize the data flow

among vehicles, and also between vehicles and data centers, ensuring efficient use of resources and enhancing the capabilities of autonomous systems.

This integration should also produce yet another benefit: energy efficiency. The heat generated by onboard computing systems can be shunted to battery systems. EV battery performance is commonly compromised by frigid temperatures; waste heat from digital electronics can be used to improve battery performance when it’s cold.

Conversely, the excess energy from the vehicle’s power system can support data processing activities, leading to an eco-friendly and energy-efficient solution.

As these technologies merge, security and privacy will remain paramount. Robust cybersecurity measures will be essential to protect sensitive data transmitted among vehicles and between vehicles and data centers. Privacy-preserving techniques will be vital in maintaining individual anonymity amidst the vast data exchange.

Enabling the trend

Power electronics and data center electronics are two engineering disciplines that have thus far remained mostly distinct. But if they are going to both be contained within a single product – EVs, their merger cannot be merely conceptual. To reap all of the potential benefits, it is necessary to integrate the two.

Building a technology ecosystem that encompasses cars, cities, data centers and communications systems that behaves optimally requires a holistic approach.

T&M is integral to every step of a product’s lifecycle, from research to design, manufacturing, conformance to safety regulations, and maintenance. Together, Tek and EA will enable a more connected world where power and data communications are integrated into a single, streamlined process.

The synergy of Tektronix and EA technologies will not only propel the industry forward, but also empower the combined team of talented engineers and scientists to address and solve some of the world’s most complex challenges in energy, communication, and engineering.

Summary

We have a big new trend that is already beginning to be transformative. A big part of this trend is actually the convergence of several others that have yet to reach their full potential, in part because they depend on the merger of the power and computing segments of the electronics industry, which is only just beginning to happen. Understandably, the ability to enable this gestalt of trends, or megatrend, is only just beginning to come together too; Tek working hand-in-glove with EA is a prime example.

—————————————————-

Denis Solomon is the automotive market segment leader at Tektronix, responsible for Tektronix & Keithley automotive solutions.

Advertisement

Stories continue below

Print this page

Related Stories