Artificial intelligence (AI) is an evolving technology that is still growing, but it is undoubtedly getting better. 

For all we know, factories might not even need lights 20 years later, because most of them will be operated by AI. We see a lot of R&D happening within the AI framework, giving good results. Besides, we see newer frameworks coming in. 

AI-based visual inspection, too, has been growing by leaps and bounds, reshaping automotive inspection. It helps examine detailed defects in vehicles, providing automotive OEMs the opportunity for accuracy and cost-savings. 

One such company focusing on AI-based visual inspection is Lincode Labs, whose AI-backed visual inspection solution, Lincode Visual Inspection System (LIVIS), is its current focus. The company was started with complete research and understanding of the top challenges that manufacturers globally face. After interviewing close to 100 customers, 86 percent of them said that quality inspection happens to be their biggest challenge.

“We were intrigued by this and went to various quality inspection people and identified the technology they were using,” says Rajesh Iyengar, Founder and CEO, Lincode Labs, and goes on, “We went on to find out that the technology hasn’t changed for two decades and there were a lot of false calls in it. That’s when Lincode stepped in and built a product around specific challenges focused on the automotive industry.”

Automotive OEMs, too, look at specifically reducing these false calls and improving productivity, which Lincode has helped solve through its AI-backed visual inspection solution. “The industry standards were 150 to 200 false calls per million inspections. So, in our case, we are doing it in zero to four parts per million,” Iyengar cites.  

Iyengar further reveals that due to this, 80 percent of their customers are repeat orders. “That’s because they are completely happy with the inspection process and the way the inspection is automated,” he mentions.  

LIVIS
Traditional vision systems cannot catch up with AI, as Iyengar says. He avers, “LIVIS stands out because we have built it as a platform. The scalability becomes easier if you’re going to deploy it across multiple factories and locations. But also, the foremost important thing is that it is completely made as a product. Thus, AI is commoditised. With the LIVIS platform, we can bring the cost and time to deployment down.” 

Lincode’s role in the automotive industry
What’s interesting is that even if Lincode caters to the manufacturing industry as a whole, it first addressed the automotive industry. The company researched the market size of various manufacturing companies and the automotive industry took the top spot, with close to USD 542 billion of global value.  

“We started with the automotive industry but pivoted later,” Iyengar tells us and continues, “So, instead of looking at just the automotive or any other industry, we turned our attention to steel, metal, plastic, glass etc. We basically went to the surface and saw how steel and metal are produced today. Whether it’s a CNC machining or forging or casting process, these are major processes used for any industry across the globe involving steel and metal. We understood that steel and metal are dealt with in the same way globally. Therefore, it made sense to go to the surface and into these kinds of defects specifically, and then generalise that and start building a model towards it. This, plus making AI as a product, has made deployment easier across the globe.” 

R&D centre in Bengaluru
Lincode recently opened a new R&D centre in Bengaluru, which also has a significant role to play in deploying the company’s solution across the world.

Stressing on the fact that evolving models are important in AI, Iyengar states, “It’s a continuous process; it’s not that you just build a model and you’re set. We have a big roadmap in the product development, and the Bengaluru R&D centre is going to play a major role in that. We are going to conduct deep research with various data collected across the globe and do various testing with that.” 

Staying ahead
What’s more, Lincode recently closed a funding round in December last year. Catering to a constantly evolving industry like the automotive, Lincode, too, strives to make sure that its visual inspection solution stays ahead and is put to use. “There are about more than 600 parts in a car and each part is segregated – like the structure, wiring, engine components etc.,” Iyengar shares and continues, “These segregations are made so that we can target the sector of the product. For example, when it comes to engine blocks, there is a specific model with a huge data set around engine blocks. This is how we stay ahead of competition.” Iyengar also adds that their trials with various use cases made them understand that inspection alone is not important but also the way the inspection is done. 

Essential skill sets for AI vision systems
Leveraging AI-based visual inspection solutions in the automotive industry is bound to increase productivity, and the cost of labour will also come down because of automation. “Today, most manufacturers use secondary inspection, which can be cut off straight away. This will improve their productivity and also reduce the risk of delays,” Iyengar enlightens. 

Moreover, AI vision systems come with their share of essential skill sets to bring out the best in the automotive industry. Iyengar states that, in general, skilling is required for the factory people. “This could be at various levels,” he puts across and adds, “It could be for the operators, the IT administrator or even the software development team. Hence, deep training is required, which can be somewhat cumbersome because it could be a bit challenging for the operator. So, an IT person might be needed in order to help the operator every time there is a downtime.” 

Covid-19 and AI-backed visual inspection
Such training or skills could certainly come in handy, because Iyengar claims that the need for AI-backed visual inspection solutions in the automotive industry has increased since the Covid-19 pandemic. “Unplanned shutdowns happened during Covid, because of which employees could not report and manufacturing could not continue properly,” he responds and adds, “Hence, a lot of investments are happening because of this. In fact, even now, a lot of employees are still not reporting and the labour problem has become global. It has become tough to get skilled workers. This has led to the adoption of autonomous manufacturing for automotives, where AI is going to play a big role.”

Meeting industry requirements
For an industry that is an economic force globally, AI-based visual inspection is certainly meeting the high-quality requirements of the customers of the automotive sector. Plus, we already see companies like Volvo using the technology. Safety surpasses any requirement, and this requirement can be fulfilled if quality is top-notch. And quality will be at its best if automotive manufacturers can perform production quality inspections in the most efficient way. (MT)

Comments (0)

ADD COMMENT

Link Copied!

California-based Drako Tech has announced DriveOS with HyperSafety, an automotive operating system designed for single-Electronic Control Unit (ECU) operation. The platform consolidates vehicle subsystems, including control systems, ADAS and digital cockpit, into one unit to reduce costs and enable over-the-air (OTA) updates.

Launched in 2015 and utilised in Drako GTE and Drako Dragon vehicles, DriveOS supports internal combustion, electric and hybrid propulsion systems.

The HyperSafety system provides real-time performance via a single-ECU architecture. According to Drako Tech, the networking backbone facilitates communication four times faster than multi-ECU Time-Sensitive Networking (TSN) automotive Ethernet.

The architecture employs hardware isolation and redundancy to maintain operation during component failures. By using a reduced code footprint and hardware partitioning, the platform aims to limit attack surfaces for cyber security and streamline validation processes.

Industry Integration

Drako Tech provides development environments that run natively on DriveOS:

• Control Systems: Allows engineers to build vehicle controls from Simulink models.
• Digital Cockpit: A system for instrument clusters, navigation and multimedia.
• ADAS: A software foundation for driver assistance with low-latency control.

The platform addresses the complexity of multi-ECU architectures, which typically require separate units for functions such as seats, doors and thermal management. Drako Tech uses a separation kernel to run safety-critical systems alongside non-critical systems, such as infotainment, on the same ECU.

DriveOS introduces hard real-time capabilities to Linux without requiring kernel changes. This allows developers to use Linux libraries and tools for safety-critical systems.

Key features include:

• Performance: 108-microsecond end-to-end performance compared to 514 microseconds for TSN Ethernet.
• Consolidation: The ability to move functions onto a single PC architecture to reduce hardware mass and complexity.
• Redundancy: Hardware-backed isolation ensures faults in one subsystem do not affect driving functions.
• Cloud Integration: Real-time fleet management and diagnostics without creating access paths to control systems.

Dean Drako, CEO, Drako Tech, said, “Nearly half of the cost of new vehicles is tied up in software and electronics. Drako Tech now offers all OEMs worldwide – regardless of size or influence – a definitive leap in their ability to deliver exceptionally safe, connected, AI-enhanced vehicles, with massive cost advantages. We are the first to achieve the ultimate goal – a single-ECU, hard real-time operating system and unified electronics architecture with mixed criticality – while providing OEMs a flexible deployment path.”

Comments (0)

ADD COMMENT

Link Copied!

German technology company dSPACE is set to showcase end-to-end test solutions at CES 2026 to assist vehicle manufacturers with the development of software-defined vehicles (SDV).

The company will present a validation portfolio featuring AI-supported software-in-the-loop (SIL) and hardware-in-the-loop (HIL) solutions.

It is exploring how generative and agentic AI technologies can support SIL testing and enable CI/CD pipelines for validation. An exhibit will demonstrate a Visual Studio Code and GitHub Copilot solution for the generation of virtual ECUs for SIL tests.

To meet the requirements of short-cycle development, dSPACE is demonstrating a CI/CT concept presenting a cloud-native validation approach. This includes a GitLab pipeline integrated with VEOS, the dSPACE SIL test software, and SCALEXIO, the HIL test platform.

dSPACE is also presenting a HIL Farm Management Demo designed to increase test efficiency. This displays the availability and utilisation of HIL systems and potential errors to reduce system downtimes and improve the use of test resources.

The technology company will use its test solutions for battery charging and battery management systems to demonstrate end-to-end SIL/HIL validation. Efficiency is increased by reusing test cases, simulation models, bus configurations and user interfaces across both methods. This allows for the demonstration of functions, including conformance tests, with the same layouts and cases.

The company is introducing DARTS ARROW, a radar solution for functional testing of sensors. Developed for end-of-line tests and periodic technical inspections, it validates safety systems such as emergency braking and lane departure warnings by simulating traffic scenarios to detect sensor errors.

For security, dSPACE will present HydraVision, a cybersecurity test framework. Using test case templates, it allows for the integration of cybersecurity tests into the development process to identify and mitigate weak points.

Additionally, the new SCALEXIO Essential system expands the SCALEXIO real-time platform. It is designed for the validation of edge ECUs for mechatronic applications in the automotive, agricultural, and construction machinery sectors. The system includes a software package and is intended as a cost-efficient entry point for HIL testing.

Comments (0)

ADD COMMENT

Link Copied!

Greaves Cotton has appointed Santosh Singh as Chief Strategy and AI Officer. He will be based in Mumbai and will lead strategy, transformation, AI-led enterprise capability building and business excellence for the Greaves Cotton Group.

Singh comes with over two decades of experience in strategy, business excellence, innovation, and AI-led enterprise transformation. He joins Greaves Cotton from Tata Technologies (TTL), where he served as Global Head – Marketing and Business Excellence. During his tenure there, he co-led the enterprise GenAI roadmap and developed use cases focused on customer engagement and productivity.

His primary mandate is to drive the Greaves.NEXT strategy, the company’s roadmap for growth across the energy, mobility and industrial solutions sectors.

In his new role, Singh will focus on accelerating growth for Greaves Technologies (GTL), developing an enterprise-wide AI roadmap, and establishing partnerships with hyperscalers and AI labs.

Parag Satpute, Managing Director & Group CEO, Greaves Cotton, said, “We are pleased to welcome Santosh to the leadership team. His extensive expertise in strategy, digital transformation, and AI will play a significant role in shaping Greaves’ next phase of growth. His global experience will further strengthen our innovation roadmap and support our long-term business priorities.”

Singh will also work across business units to incubate and scale growth vectors and lead business excellence initiatives.

Comments (0)

ADD COMMENT

Link Copied!

Luminar Technologies, Inc., a global technology company advancing safety, security and autonomy across various sectors, has announced it has agreed to sell its wholly owned subsidiary, Luminar Semiconductor, Inc. (LSI), to Quantum Computing Inc. (QCi) for USD 110 million in an all-cash transaction.

QCi is an integrated photonics and quantum optics technology company focused on photonics-driven technologies and sensing applications. LSI's innovation platform and engineering depth align with QCi’s strategic priorities in optical systems, chip-scale innovation and photonic architectures. The acquisition is expected to position LSI to grow and capitalise on the demand for photonics solutions.

Paul Ricci, CEO, Luminar, said, "We are pleased to partner with QCi as they continue to accelerate their photonics roadmap. QCi’s focus on photonics-driven technologies provides an aligned platform for LSI to expand its customer base, accelerate growth opportunities, and invest in markets where long-term demand for high-reliability optical systems is increasing. We are incredibly proud of the LSI team for the progress they have made to reach this milestone, and we are excited for the opportunities ahead for LSI under QCi’s ownership.”

Yuping Huang, CEO, Quantum Computing Inc, said, "I’m excited about the opportunity to partner with the exceptional team and valued customers of LSI. There is clear strategic alignment and shared vision between our organizations, creating strong momentum from day one. Following the closing, we will move quickly to invest in and scale LSI’s existing business, while bringing our teams together to accelerate our quantum photonics roadmap. This is a powerful combination, and I’m energized by what we will achieve together.”

In a separate announcement, Luminar announced that it has initiated voluntary chapter 11 cases in the U.S. Bankruptcy Court for the Southern District of Texas. LSI is not a debtor in the chapter 11 cases and is operating in the ordinary course. Because LSI is a subsidiary of Luminar, the transaction will require the approval of the bankruptcy court via a Section 363 sale process, which the parties expect to receive by the end of January 2026, subject to closing conditions.

Comments (0)

ADD COMMENT

Link Copied!