According  to its 2025 MEMS market report, Bosch remains the leader in MEMS, with revenues of $2bn in 2024, with a year-on-year increase of 12%. Bosch benefited from the end of the inventory crisis and the robust performance of its smart sensor portfolio, which integrates intelligence to deliver value beyond basic sensing. Its strong positioning in high-end systems that were less affected during downturns has also shielded it from deeper volatility, says Yole.

John Li, Technology Analyst at IDTechEX

Autonomous emergency braking (AEB) is one of the most important ADAS (advanced driver assistance systems) features a vehicle can have. Studies have shown that pedestrian crash risk is reduced by 25-27 per cent and pedestrian injury crash risk by 29-30 per cent through the implementation of AEB. As discussed in IDTechEx’s new report, “Infrared (IR) Cameras for Automotive 2025-2035: Technologies, Opportunities, Forecasts”, LWIR (long-wave infrared) camera technologies could drive the improvement of AEB systems in darkness and adverse weather conditions. IDTechEx assesses existing regional regulations on AEB and the effect they will have on the adoption of infrared camera technologies in ADAS.

The US passenger vehicle market will be the first to adopt LWIR cameras due to NHTSA’s US regulations

The first and most important regulation to be aware of is NHTSA’s ruling for AEB and pedestrian AEB in the US. In its report, NHTSA details the progress made in vehicle safety due to AEB, before setting a 2029 deadline for all new vehicles installed to have AEB with more stringent requirements. A key requirement is that all new vehicles pass automatic braking tests against lead vehicles and pedestrians using a minimum level of illumination (low-beam headlights) with ambient lighting of 0.2 lux (night), as over 75 per cent of pedestrian injuries occur in non-daylight conditions.

NHTSA believes that only 5 per cent of new vehicles by 2029 will require hardware upgrades to meet these demands, as most vehicles will have advanced camera and radar technologies by 2029, such that only software upgrades are required. However, the more pessimistic response from automotive OEMs and tier-one suppliers is a suggestion that this is potentially optimistic.

While current AEB systems use a combination of camera and radar technologies, both SWIR (short-wave infrared) and LWIR technologies would provide an additional sensing mode that is unaffected by darkness or fog. IDTechEx predicts that some OEMs will turn to these technologies to meet NHTSA’s requirements, while others will upgrade existing camera and radar hardware, or may install additional hardware such as LiDAR. Further details, analysis, and forecasts are discussed in IDTechEx’s new report, “Infrared (IR) Cameras for Automotive 2025-2035: Technologies, Opportunities, Forecasts”.

Opportunities and regulations in Europe, China, and other regions

While regulations in other regions haven’t been as explicit yet, there are targets regarding AEB and vehicle safety overall, which may have an impact on the uptake of AEB in vehicles. In the EU, all new vehicles were mandated to have AEB systems installed in 2022, and this was extended to existing vehicle models in 2024. On top of this, the EU’s Vision Zero initiative aims to achieve zero road fatalities by 2050, with an intermediary goal of halving the 19,800 fatalities in 2021 by 2030. This included a ~US$7.6 million project, PROSPECT, which investigated camera, radar, and LiDAR technologies for pedestrian detection and AEB.

In 2023, the number of road fatalities in the EU was approximately 20,400, only a 1 per cent decrease compared to the number of road fatalities in 2022. This includes countries such as Spain, France, and Italy, where fatalities have barely decreased since 2019, and Belgium, Hungary, Poland, and others, where the number has increased. Combined with Europe-centric projects such as the HELIAUS Project and the lack of progress in reducing pedestrian fatalities, IDTechEx expects the technology developments driven by NHTSA’s regulation to also be adopted in Europe from 2027, although this will be more gradual due to the lack of concrete regulation.

The market penetration of AEB in China lags behind Europe and the US overall. IDTechEx estimates that the market penetration of AEB in the Chinese passenger market is approximately 40 per cent as of 2024. In terms of sheer volume, however, China sold 30.1m passenger vehicles in 2023, compared to 10.5m in Europe in the same year. Combined with the projected increase in AEB penetration to exceed 60 per cent by 2030, China presents a market opportunity for LWIR thermal cameras over the next 10 years. As is expected in Europe, the technology developed in anticipation of the US market will result in a rise in LWIR-fused AEB in the Chinese market, starting from 2027.

The three largest passenger vehicle markets (the US, Europe, and China) all have their own New Car Assessment Programs (NCAP). These are evaluations of a passenger vehicle’s overall safety using standardized test conditions. In the US, these are operated by NHTSA and are mandatory for every new vehicle, including its new night-time AEB condition tests, to be carried out in 2029. The Euro NCAP, in contrast, is voluntary but has its own tests for vehicle and VRU (child/adult pedestrian, cyclist, motorcyclist) encounters. At the start of 2024, China’s new C-NCAP regulations replaced the previous version from 2021 and placed greater importance on ADAS features such as AEB and VRU detection.

The C-NCAP involves AEB tests with pedestrians and two-wheelers, with speeds ranging from 10-80 km/h. It also covers scenarios where pedestrians are hidden from view or when a vehicle is making left and right turns. IDTechEx is aware of two Chinese OEMs introducing LWIR night vision into their vehicles. The technology is on the roads from these OEMs, and also OEMs such as General Motors, BMW, and Mercedes, which have historically had optional night-vision systems.

Hardware and software upgrades would be required to integrate LWIR sensing into AEB for ADAS, but the fact that the technology is already being used reinforces IDTechEx’s forecast for LWIR-integrated AEB to enter the Chinese market at a similar time and rate to in Europe. Other regions, including Japan and India, have their own NCAP-style tests for assessing the overall safety of vehicles.

Autonomous driving and IDTechEx outlook

AEB will encourage the adoption of LWIR cameras in SAE level 1-3 vehicles, with most current on-road vehicles being level 1 or 2 in the European, US, and Chinese markets. IDTechEx predicts that there will be an even larger demand for LWIR cameras for autonomous driving when SAE level 4 vehicles and robotaxis enter regional markets in the long term.

With greater demands on sensing for human-less driving put on thermal sensors, this is where higher definition cameras, as well as upgraded software and stereo vision technologies, would be required of LWIR cameras. From the early 2030s, IDTechEx expects autonomous driving in SAE level 4 vehicles and robotaxis to be the dominant driver to further LWIR camera adoption in passenger vehicles. Granular forecasts by region and SAE level can be found in IDTechEx’s new report, “Infrared (IR) Cameras for Automotive 2025-2035: Technologies, Opportunities, Forecasts”.

IDTechEx expects the initial uptake of automotive LWIR cameras to increase the yearly market size by 6.5 times between 2025 and 2030, driven primarily by NHTSA’s regulations, as well as updated regional NCAP testing programs and safety initiatives. IDTechEx expects SAE level 4 vehicles and robotaxis to enter the market at the start of the 2030s. This will further drive automotive LWIR camera adoption, increasing the market by 2.5 times from 2030 to 2035. LWIR cameras for autonomous driving will require higher resolutions and possibly more per vehicle, compared to LWIR cameras for AEB. Source: IDTechEx – “Infrared (IR) Cameras for Automotive 2025-2035: Technologies, Opportunities, Forecasts”

To find out more about this new IDTechEx report, including downloadable sample pages, please see www.IDTechEx.com/InfraAuto.

For the full portfolio of sensors market research available from IDTechEx, please visit www.IDTechEx.com/Research/Sensors.

Automotive cyber incidents are increasing year on year, with rising hacker activities being the prevalent factor.

Another clear trend is the growth of remote hacks, including both web-based and nearby wireless attacks. However, the challenge of defending against these attacks is further compounded by ongoing issues associated with interpreting and understanding cybersecurity jargon. As connectivity and software-driven processes have become ubiquitous across the industry, there is a clear and present danger that failure to address this issue will mean that privacy – and even consumer safety – is actively being compromised. 

With automotive leaders challenged with defending against sophisticated adept and opportunistic cybercriminals, access to understandable and actionable threat intelligence is a must-have tool to support all businesses. However, the automotive c-suite is struggling to connect the real implications of their threat intelligence to specific business operations, with almost a third (29.5%) of respondents not seeing value from their cyber intelligence investments. 

Increasing attacks

For a problem that touches everyone, from top to bottom in any organisation, cyber literacy is a critical component if an increasingly interconnected automotive industry is to develop a culture of cybersecurity best practice, share knowledge and, ultimately, institute actionable intelligence with clear and quantifiable return on investment. Failure to move on from tactics that rely on reactive security and risk management principles without truly understanding the risks will provide inadequate security protections, exposing organisations to unnecessary and significant cyber risk and the consequences of devastating intrusions across the entire network.

Our findings suggest that, for automotive businesses to start seeing true value from their threat intelligence investments, they should start viewing the challenge more holistically. Protecting business operations whilst tackling cybersecurity threats has radically changed – from basic IT configurations, installing an antivirus and following best practices, to a whole new level of complex coding, unknown threats and continuous cyberattacks.

Good intelligence reports and timely warnings are critically important for automotive companies across the supply chain as cybercriminals increasingly turn their focus towards the automotive industry.

By Clara Wood, Automotive Research Leader, Kaspersky

Test and measurement giant Tektronix late last year commissioned a global survey of some 1,300 T&M engineers to discover what are their current requirements and concerns when using T&M equipment. The results of the survey are out and show the following trends: To keep pace with rapid advancements in technology, today’s test and measurement engineers seek longer lasting products and more technical support

• Long instrument lifetime and greater reliability

Nearly 40% of the survey respondents expect equipment to last more than ten years, whilst a fifth (some 21%) demand it to last over five years, backing up findings that engineers place a high value on instrument reliability.

The survey reveals that engineers want to protect their investments and get continued value from their purchases, which is in line with T&M vendors who continue to offer service and technical support. Tektronix is among those that offer different warranty lengths according to product type. At Tektronix, calibration and repair support for each product typically continues for a minimum of five years after a product has been discontinued.

• Technical support

Among the services engineers value the most is technical support – over 70% of the survey respondents require it.

Tektronix strives to support engineers’ needs by offering evolved ways of working, including remote control capabilities with its TekScope and TekDrive, collaborative T&M data workspaces that enable cloud-based software-defined engineering – this is trend that grew significantly in importance over the last two years. Throughout the pandemic, solutions such as virtual online product demonstrations have shown to be very popular with T&M customers.Tektronix also provides its customers with support through a dedicated technical Pan-European support team and a European multilingual ‘Center of Excellence’ (CoE) team.

• Post-sales support

Post-sales support such as calibration, repair and training were also listed to be of significant value to today’s engineers.
Tektronix has one of the world’s most comprehensive networks of repair and calibration services for any brand of test and measurement equipment, and also offers asset management and factory service plans.

Extensive online training resources, including application notes, training videos and webinars, all helps today’s engineer.

• Other requirements

Other key factors that survey participants said they value from their vendors include:

• Comprehensive, easily-accessible and quick-to-navigate user manuals (69%),
• A good supply of accessories and probes (50%), and
• Extensive and detailed application notes (43%). 

The last few years have seen a massive uptake in businesses prepared to lead the way in sustainability. Some 90% of the largest 500 companies by market cap in the Russell index published sustainability reports in 2019. This is driven by a range of new trends including legislation such as the UK Environmental Bill, Mandatory Climate Disclosures (which will require companies to report on their GHG emissions) and November’s COP26 – all creating a new sense of urgency.

There is also pressure from within companies, from shareholders and even employees; for example, Amazon employees joined the climate strike. Perhaps the biggest push is from customers (B2B and B2C) who are becoming environmentally-conscious buyers. Research shows 79% of customers say they are switching to businesses that promote social responsibility and sustainability.

In response, many electronics manufacturers are now promoting their products as “sustainable”, “environmentally-friendly” and other vague terms. This can help boost sales, but what happens when these claims are not backed by a real change, i.e., all talk, no action? Given there is such a huge incentive for businesses to present themselves as green and sustainable it should not come as a surprise that not all of these claims are backed by evidence of improvement. This is referred to as “greenwashing” – misleading consumers about a product or brand’s environmental credentials. In fact, the problem has become so bad that the UK’s Competition & Markets Authority (CMA) has released their anti-greenwashing code to tackle it.

The CMA found that 40% of green claims could be misleading. A similar effort is underway in Europe. The European Commission found that 42% of claims to be “green” were “exaggerated, false or deceptive with 59% failing to provide evidence”.

In 2021, issuing press releases and media statements about environmental commitments is not enough, and consumers are fast catching on. Nearly 50% of consumers say they do not have any information to verify retailers’ sustainability claims and, worse yet, 44% say they do not trust product sustainability claims.

There is also a range of issues in today’s electronics space that creates cynicism such as, planned obsolescence, making devices difficult to repair (cue in Apple products), and the upgrade cycle where technology becomes quickly outdated and devalued. These issues create tension even towards bold sustainability claims.

Ultimately, making unfounded claims can actually damage a business, from its reputation to even falling foul of the CMA’s new anti-greenwashing code.

This latest crackdown by the CMA and similar efforts in Europe are encouraging. It will only benefit manufacturers who are taking steps to improve their circular economy, reduce waste, cut down on packaging, and reducing emissions.

So, how can a company ensure it delivers on its sustainability promises? It is down to the evidence. Being able to scientifically prove and demonstrate a sustainability approach at all levels of the organisation is a way for businesses to meaningfully engage with their customers. This will create brand preference and even improve business efficiency and reduce costs. To achieve it, businesses need real-world evidence and data: evidencing improvements, testing new approaches, and establishing the practices that work.

“We have been working with some amazing companies who are desperate to get their hands on scientific data so they can start making a real impact. We are giving businesses opportunities to reduce pollution, identify what product is causing it, and our customisable reports easily and scientifically evidence their progress,” said Freddie Talberg, CEO and founder of EMSOL.

[Image: Michael Dziedzic for Unsplash]

Power electronic devices are vital to the operation of nearly every device that runs off electricity, yet they are an underappreciated segment of the semiconductor space.

Generally, they are found in power converter circuits, to convert AC from the grid to DC voltage, or between DC voltages, or run electric motors (DC to AC conversion), and more. Their continuing improvement has increased the efficiency of every electronic device, and are today considered crucial to enabling energy savings and carbon-reducing technologies such as LED lighting, solar power generation and electric vehicles.

Suitable materials

Regularly found in power electronics today are semiconductor materials are the so-called wide-bandgap (WBG) materials, Silicon Carbide (SiC) and Gallium Nitride (GaN). They gained their name due to their electron energy bandgap is wider than silicon’s, which gives them very beneficial characteristics such as lower electrical resistance and higher frequency switching than the current power-workhorse devices IGBTs and MOSFETs.

Although often grouped together, in reality, SiC and GaN have important differences between them, with separate “sweet spot” where the material is most suitable.

Wafer sizes are the first big difference: SiC ingots are grown from a single-crystal seed wafer by chemical vapour deposition (CVD) or physical vapour transport (PVT). Both methods require high temperature and are slow compared to creating silicon ingots. SiC wafers also require slicing the ingot into disks. Since this is a very hard material and difficult to cut – even with a diamond saw, it proves challenging. There are several methods to do ingot separation, but they can introduce defects into the single crystal.

In contrast, GaN substrates are not cut from ingots, but are grown by CVD on top of silicon wafers, which in itself is a challenge since there’s a lattice constant mismatch between the two materials. There are methods to engineer this strain, but there is a risk of defects which will affect reliability. Because GaN sits on top of the silicon, GaN power devices are considered “lateral”, i.e. their source and drain are on the same side of the wafer. – in contrast to silicon and SiC power switches, where the main current path is vertical, through the chip.

Different strenghts

SiC and GaN also have different voltage levels where they are optimal. GaN devices rated for breakdown voltage of about 100V will find uses in mid-voltage power conversion from 48V down, with applications in cloud computing and telecom infrastructure. Additionally, power supplies and wall warts will contain 650V GaN power switches, which is the right voltage rating for AC to DC conversion with the wide input voltage range of 90-240VAC. The high frequency of GaN allows the passive components of the power supply to be significantly smaller, resulting in overall much more compact solution.

In contrast, SiC devices are designed for 650V and above. It is at 1200V and higher that SiC becomes the best solution for a many applications, such as solar inverters, electric vehicle chargers and industrial AC to DC conversion. Another long-term application is the solid-state transformer, where the current copper and magnet transformers are replaced with semiconductors.

Power revolution

The next revolution in power electronics is already upon us. The up-and-coming wide-bandgap materials of silicon carbide and gallium nitride will help make the future of power electronics more efficient and more compact for a wide variety of applications.

We see developers and providers of products focusing on improving communications solutions for our remote working world. Hence, the five trends we see for 2021 are as follows:

1. Video communications

Often called videoconferencing, video communications, may appear as old news. But, 2020 delivered an unimaginable punch and sent much of the world’s population to work from home. Videoconferencing became the saviour, allowing a degree of normality for hundreds of millions of people.

Videoconferencing capabilities offered today by Zoom, Microsoft Teams, Google Meet, GoToMeeting and others are quite formidable solutions compared to what was available five years ago – ease of use, high quality audio and video, reliability and virtual backgrounds are just some of the features on offer. Yet, the journey is far from over.

Videoconferencing is still a focused activity, requiring preparation, proper setting, scheduling and more. In 2021, expect videoconferencing to become easier to use and more secure. The widespread use of videoconferencing might even bring back some of the old research projects – for example, the EU-sponsored AMI project, where metadata was established during the call to fully analyse the dynamics of conversation for dominance and other essential characteristics.

Voice UI, better presentation capabilities similar to those offered by the mmHmm startup, and AR/VR/MR might be the subject of widespread research and innovation throughout 2021. We may also see some companies building virtual Walls of Presence similar to the one Orange (then France Telecom) enacted between offices, allowing people to meet for coffee without having to travel.

Star Trek’s Holodeck might become part of every house one day and whilst we overshoot 2021 by a few good years in mentioning this, it serves to emphasise that video communications improvements and better functionality are definitely trending and will continue at pace.

Industries across the world have seen remarkable gains in employee productivity this year through home and virtual conferencing, and 2021 may well be the year when it becomes the permanent norm.

• Virtual events

Physical exhibitions like CES, MWC and IWCE are not gone forever. The real handshake is still important, but it will have to wait for the time being. As videoconferencing allows industry executives to “meet” for most of 2021 – and possibly beyond – so exhibition organisers must embrace the virtual world to survive. This, however, could be easier said than done. Physical attendees are much easier to engage once at the event, a format more difficult to replicate virtually. We should see much innovation in this space in 2021.

• Virtual banking

Virtual banking is yet another “old news” topic – or is it? Even in the pre-pandemic days, video had great potential for the banking industry, especially for operations in rural areas. Now that video is universally acceptable, and banking is often done via the mobile phone, adding video to banking applications is an obvious trend. But, exacerbated by the pandemic, new areas will start taking advantage of virtual banking capabilities – look at the mortgage industry as an example. You still need to guarantee security and privacy, and verify and accept electronic signatures – this space is ripe for innovation.

• Telehealth

Since the pandemic struck, telehealth quickly evolved to become the best tool for healthcare practitioners to help patients – without the risk of spreading virus. However, current telehealth capabilities are still very limited in standard patient visits and consist mainly of a basic conversation. The capabilities of today’s smartphones and smart watches in terms of obtaining a variety of health care data at any time, is already impressive and should be coupled with emerging telehealth applications in transmitting health data measurements, analysis, alerts and more.

• Automatic translation

Automatic translation solutions now allow books, articles and websites online to be translated reasonably fast and cheaply. The next development is having a little box you can speak into, which will instantly translate your speech into the language of choice. Imagine a conference call where different-language speeches are translated and captioned in real time. Automatic translation could feature in telehealth and virtual banking apps, opening up endless possibilities for innovation.

[Image: Mika Baumeister for Unsplash]

As the European Union implements its Green Deal goal to be carbon neutral by 2050, governments and environmental groups are demanding cleaner cities and a reduction in greenhouse gases and other pollutants. To stimulate demand, European countries are increasing subsidies to vehicle owners to encourage them to invest in electric vehicles as a reliable and alternative clean method of transportation. This change is in turn driving manufacturers of vehicles as diverse as e-bikes, cars, vans, buses, agriculture and construction equipment to rethink their design and to use electric and hybrid transmissions to increase the efficiency of their vehicle powertrain whilst eliminating exhaust gases and pollutants.

For the engineers developing these next-generation systems, selecting the right high-voltage connection system can be a daunting task. Should they use shielded or unshielded connection systems, plastic or metal connectors, flat blade or circular contacts, PCB or busbar-mounted connection systems? How will they meet their sealing performance, vibration and latching requirements? How can they reduce cost in their systems, and are there second-source suppliers available? These challenges are becoming increasingly difficult as manufacturers introduce second- and third-generation versions of their high-voltage connection systems.

In automotive applications, where production volumes are usually high, engineers are trying to find a balance between performance, reliability and cost. Connection systems are usually manufactured in plastic housings and feature HVIL contacts, TPA and CPA components as standard. Most connectors support voltage levels up to 800V, with many German OEMs insisting that these connectors are designed and tested according to LV214 and LV215 standards whilst other OEMs require validation acc. USCAR or an OEM-Specific standard. Many connectors feature flat blade contact systems that both increase the current-carrying capability and enable reliable crimp termination whilst maintaining a competitive cost level. Examples include TE Connectivity’s PCON12 terminal system, which is used in the new HVA1200 AK Class 2/3 product family that supports currents up to 100A for auxiliary device applications.

Cost is another critical topic and some OEMs are demanding second sources for their connection systems. Suppliers like Aptiv are therefore now offering products that are 100% intermateable with other AK Class 1, AK Class 2, AK Class 4 and AK plastic Pass-Through connection systems, whilst simultaneously improving the connector sealing and vibration performance, with respect to the existing designs. Aptiv has also developed its new HV RCS 800/890 Direct Mate connection system, where the cable plug connects directly to metal device housing, again removing cost from the overall interface.

An associated trend is that vehicle manufacturers are reverting back to using lower-cost unshielded cables and connection systems in combination with EMI filters to again remove cost and weight from their systems. In line with this trend, other OEMs are evaluating moving from copper HV cable to an aluminium option in the mid-range amperage and voltage ranges. There is still a long way to go before proliferation, and there are a lot of considerations to be aware of, including increased diameter size, cable flexibility and product availability, amongst others.

In heavy equipment and commercial vehicle applications, designers are focusing on the challenges of meeting the higher power requirements of these vehicles. Whereas truck and bus OEMs still tend to prefer plastic connection systems, agriculture and construction vehicle OEMs often prefer to use metal connector systems for increased ruggedness and reliability. Voltage specifications of up to 1000V are considered standard and current levels above 500A are becoming more common. This requires customers to use thicker 70 mm², 95mm² or 120mm² cables, which can be both heavy and relatively expensive. Finding a reliable contact system that allows users to limit cable sizes can help optimise design cost.

A final design consideration for engineers is their high-voltage cable assembly strategy. High-voltage cables are challenging to assemble, and the required crimping and testing equipment can be expensive, so not all traditional harness makers will be able to produce these cables. Particularly for small-volume OEMs, engineers will need to consider up front if will they build cables in-house, source via a local cable harness maker or distributor, or if their volumes will be sufficient to be supplied directly by a connector manufacturer. Finding the right local partner or distributor during the development phase can make sourcing a lot simpler when their vehicle is ready to go into full production.

As the European Union implements its Green Deal goal to be carbon neutral by 2050, governments and environmental groups are demanding cleaner cities and a reduction in greenhouse gases and other pollutants. To stimulate demand, European countries are increasing subsidies to vehicle owners to encourage them to invest in electric vehicles as a reliable and alternative clean method of transportation. This change is in turn driving manufacturers of vehicles as diverse as e-bikes, cars, vans, buses, agriculture and construction equipment to rethink their design and to use electric and hybrid transmissions to increase the efficiency of their vehicle powertrain whilst eliminating exhaust gases and pollutants.

For the engineers developing these next-generation systems, selecting the right high-voltage connection system can be a daunting task. Should they use shielded or unshielded connection systems, plastic or metal connectors, flat blade or circular contacts, PCB or busbar-mounted connection systems? How will they meet their sealing performance, vibration and latching requirements? How can they reduce cost in their systems, and are there second-source suppliers available? These challenges are becoming increasingly difficult as manufacturers introduce second- and third-generation versions of their high-voltage connection systems.

In automotive applications, where production volumes are usually high, engineers are trying to find a balance between performance, reliability and cost. Connection systems are usually manufactured in plastic housings and feature HVIL contacts, TPA and CPA components as standard. Most connectors support voltage levels up to 800V, with many German OEMs insisting that these connectors are designed and tested according to LV214 and LV215 standards whilst other OEMs require validation acc. USCAR or an OEM-Specific standard. Many connectors feature flat blade contact systems that both increase the current-carrying capability and enable reliable crimp termination whilst maintaining a competitive cost level. Examples include TE Connectivity’s PCON12 terminal system, which is used in the new HVA1200 AK Class 2/3 product family that supports currents up to 100A for auxiliary device applications.

Cost is another critical topic and some OEMs are demanding second sources for their connection systems. Suppliers like Aptiv are therefore now offering products that are 100 % intermateable with other AK Class 1, AK Class 2, AK Class 4 and AK plastic Pass-Through connection systems, whilst simultaneously improving the connector sealing and vibration performance, with respect to the existing designs. Aptiv has also developed its new HV RCS 800/890 Direct Mate connection system, where the cable plug connects directly to metal device housing, again removing cost from the overall interface.

An associated trend is that vehicle manufacturers are reverting back to using lower-cost unshielded cables and connection systems in combination with EMI filters to again remove cost and weight from their systems. In line with this trend, other OEMs are evaluating moving from copper HV cable to an aluminium option in the mid-range amperage and voltage ranges. There is still a long way to go before proliferation, and there are a lot of considerations to be aware of, including increased diameter size, cable flexibility and product availability, amongst others.

In heavy equipment and commercial vehicle applications, designers are focusing on the challenges of meeting the higher power requirements of these vehicles. Whereas truck and bus OEMs still tend to prefer plastic connection systems, agriculture and construction vehicle OEMs often prefer to use metal connector systems for increased ruggedness and reliability. Voltage specifications of up to 1000V are considered standard and current levels above 500A are becoming more common. This requires customers to use thicker 70 mm², 95mm² or 120mm² cables, which can be both heavy and relatively expensive. Finding a reliable contact system that allows users to limit cable sizes can help optimise design cost.

A final design consideration for engineers is their high-voltage cable assembly strategy. High-voltage cables are challenging to assemble, and the required crimping and testing equipment can be expensive, so not all traditional harness makers will be able to produce these cables. Particularly for small-volume OEMs, engineers will need to consider up front if will they build cables in-house, source via a local cable harness maker or distributor, or if their volumes will be sufficient to be supplied directly by a connector manufacturer. Finding the right local partner or distributor during the development phase can make sourcing a lot simpler when their vehicle is ready to go into full production.

The application software sector is undergoing profound and rapid change. In our report “Tech, Media, & Telecom Trends 2020 – Thematic Research” we reveal how historically the value was in applications but this year it is driven by the integration of software.

Here are the top application software technology trends we’ve identified for 2020:

• Cloud: The cloud is an important model for IT resource delivery. The growth of application programming interfaces (APIs) has made sharing data easier and encouraged the development of new applications. A new wave of platform services will use artificial intelligence (AI) to increase operational efficiency and automate workflow controls. Advanced platforms will enable intelligent apps and improve user experience.

• Process automation: Process automation refers to robotic process automation (RPA) in which new developer technologies will help shore up various phases within the application lifecycle. RPA uses AI, low-code platforms and software robotics to automate manual processes. Industrial IT providers will continue to invest in RPA innovation.

• Machine learning: Machine learning is a high priority for businesses everywhere. There are relatively few AI specialists, and developing a model is a time-consuming and resource-intensive process. As the market matures, enterprises will incorporate data from a variety of sources into their machine-learning models.

• AI libraries: To overcome the skills gap, vendors are designing products specifically for people without data science skills. Vendors are offering translation services, speech-to-text, natural language generation, image recognition and video analysis to non-AI specialists.

• Data will become a key differentiator, and an enterprise’s ability to manage and glean insights from it.

• Conversational platforms: Conversational platforms employ a variety of technologies to enable human-like interactions with computer systems. Virtual personal assistants like Amazon’s Alexa or Google’s Assistant can schedule appointments, provide weather updates and play music based on voice commands. In 2020, application platform providers will develop user interface tools to support advanced conversational platforms.

• Open-source AI: AI relies heavily on open-source technologies. Google’s machine-learning framework, TensorFlow, is a prime example of this trend. Released via the Apache OS license in 2015, TensorFlow is currently available to developers across most public cloud platforms. Investment in open-source AI technologies will proceed apace, particularly among those companies selling both open and proprietary technologies.

• Augmented reality (AR): AR allows the user to see the real world overlaid with a layer of digital content. Having come to worldwide attention in 2016 with Pokémon Go, the next two years will be pivotal for AR. Its chances of crossing over to the mass consumer and corporate markets will be determined by the richness and variety of AR content, the development of smartphone applications and the arrival of all-in-one glasses.

In a recent XMOS-undertaken survey of design engineers, 56% said their on-device processing requirements are already high or very high, and another 40% need even more on-device processing. But, 38% claim the goal is to reduce the bill of materials, and of these 48% claim it’s difficult to achieve, because of higher power consumption and greater design complexity.

Artificial intelligence (AI) at the edge is the latest significant demand that is pushing these design boundaries. Market analysis firm Gartner forecasts the AIoT market to reach $3trillion by 2025. In addition to existing compute, the AIoT demands a whole new class of processing across a broad range of applications. The tension is obvious — a solution that does not deliver the required features is irrelevant, if it is too expensive it will fail to be adopted (or the resultant product will fail in the market), if it is too complex it may miss the market window.

If process technology is no longer the answer, then we must address the systems that we build on silicon. Chip vendors should consider how their technology can be re-imagined to deliver on the needs of a new generation of smarter devices. These considerations should include the architecture of their device portfolios, its implication on ease of use and, of course, cost. They should also consider the future — how do their solutions scale with the relentless increase in demand for more capabilities?

All this means is that system architectures must now support four classes of workload — control, signal processing, IO and communications, and AI. Within this in mind, let’s consider some primary factors that drive our choices of architecture:

The development costs of a semiconductor must be amortised over the volume shipped in its lifetime. It follows that, all other things being equal, a semiconductor that can target a large market will be cheaper than one that can target a smaller market. Accordingly, as customer requirements diversify and the market for a specific set of features decreases, the cost of a chip delivering that fixed set of features must increase.

Conversely, a flexible device that can address a broad set of applications can maximise the return on R&D investment and those savings can be passed on to the customer, reducing cost. As diversity increases, it becomes better to spend R&D dollars on optimising flexible technology than increasing the number of application-specific devices.

Homogeneous or heterogeneous?

The right approach is homogeneous for as long as it is possible to deliver on required performance and cost. To make an analogy, if I have a two-seater sports car and an SUV, I can deliver on speed or luggage space, but I cannot deliver on both at the same time. A homogeneous solution would be to have two fast SUVs. Obvious, but what of efficiency? Well, both fast SUVs could probably be smaller, maybe not individually as quick — dealing with peak demands by working together.

It is only when we cannot otherwise accommodate the performance or cost demands of the application that we should consider heterogeneous approaches — they are inflexible and more complex to program.

One development method or four?

As we enter into the new age of AIoT because it is likely that different techniques will be used by the embedded programmer, the DSP engineer and the data scientist. However, the guiding principal should be to support industry standard approaches that are common in the customer and partner ecosystems, and that those approaches should ultimately be consolidated into a single tool flow.

The more homogeneous the target device, the easier this is to achieve. Homogeneous devices also scale elegantly. If you need more resource, more of the same resource is added. Done correctly, the design complexity impact of scaling a homogeneous architecture should be negligible. In the end, an easier to use architecture gets designs into the market faster.

Power electronics is a growing market thanks to electrification of many industries including the automotive industry. The power electronic modulus are also growing in performance. In particular, the are becoming smaller, lighter weight, more tightly integrated, and better able to handle higher power levels.

In some cases the semiconductor technology is also shifting from Si IGBT to SiC MOSFET. The dies are shrinking in area and the number of push-pull pairs in the circuit are being reduced. All these translate into higher power densities and higher operating temperatures, as outlined in the IDTechEx Research report “Conductive Ink Markets 2019-2029: Forecasts, Technologies, Players”.

In many cases, the bottleneck today is the material used in the power module package. The aluminium wire bond and the die and substrate attach are common modes of failure. The high operating temperatures push performance requirements beyond the capabilities of SAC solders. This opens the door to alternatives.

In particular, metal sintering has emerged as a promising high-performance and high-cost alternative. The most common version is based on micron-sized Ag particles and requires pressured sintering. The trend now is to improve pressure-less sintering which is traditionally constrained to small die areas and suffers from long sintering times. To this end, nano Ag or hybrid (nano and micro) sinter particles are developed. Furthermore, there is today innovation in form factor, going beyond pastes to offer dry films and dips to make it easier for the user to adopt. The metal sintering technology is now qualified in the automotive sector after 5-7 years of development.

Cu versions are also in competition. As usual, the promise is the same: offer same or higher performance than Ag whilst cutting costs. The early results are promising but as usual Cu solutions are not a drop-in replacement since their sintering conditions are different. This year and next year will probably be make-or-break years for Cu alternatives as serious testing is well underway.

In general, metal sintering is high performance. The thermal conductivity is high and the melting temperature of a sintered die attach line is comparable to a solid metal, thus offering very high homologous temperatures. The cost however is still a significant barrier. This is both direct material costs and also the associated costs in terms of new equipment, new required surface finishes/metallizations, and so on.

This technology is however now on the move. The number of suppliers has multiplied. Almost all module makers have worked with and prototyped using metal sintering solutions. As such, we assess that the future of this market as highly promising.