Driving intelligent automotive manufacturing: how robotics and AI are transforming OEM operations

Robotics in automotive: a major growth area for the next decade

Robotics is generating headlines in the automotive industry right now, with major investments expected over the next few years. The market for robotics in automotive could grow from around $9bn in 2024 to around $22.5bn in 2033, according to one source.

Why is there so much excitement now, when robots are already firmly established in automakers’ factories? To a great extent, the upward investment trend is the result of transformational opportunities presented by the integration of robotics with emerging technologies.

In general, operational technology (OT) is being increasingly automated through the application of machine learning (ML) and AI, including agentic AI. (For more about this topic, please read our POV on the new AI imperative in manufacturing, co-authored with Microsoft.)

One of the results is a new generation of more intelligent robots. A key concept here is physical AI, where AI is combined with machines so that it becomes able to interact with the physical world. This paves the way for adaptive processes that improve efficiency, precision, and safety.

This concept can revolutionize automotive production lines, making it possible to automate key tasks such as quality control. For example:

• Quality inspection: AI-driven vision systems can use high-resolution cameras and deep learning to inspect vehicles and components with superhuman precision.
• Defect detection: AI-driven robots can identify microscopic surface flaws, misalignments, or paint inconsistencies that are difficult for human inspectors to spot.

Because physical AI results in autonomous systems that perceive, understand, and interact with the world, it has a major part to play in intelligent manufacturing.

Adding AI to existing robots

Of course, automotive factories are already full of robots, and it will not be commercially viable to replace these with next-generation models equipped with the embedded chips needed to run AI. Nor do most existing robots have the capacity to have AI retrofitted.

Instead, AI capabilities can be added at the edge, and can then interact with the robots to give them advanced capabilities such as context awareness. This type of edge AI / robotics suite, along with hybrid edge-to-cloud platforms, is evolving rapidly and becoming highly relevant to automotive. Digital twins are increasingly used for validation of the resultant features.

Benefits of advanced robotics

This integration and convergence will position automakers to improve productivity and cut costs, strengthening their leadership in the current high-pressure business environment. That is mainly because the addition of technologies such as AI and ML to robots will give them the ability to take over a whole range of tasks that were previously hard to automate because of their unpredictability.

As MIT’s Daniela Rus explained in a recent conversation with Capgemini: “In manufacturing and logistics, robots will no longer be limited to repetitive tasks. They’ll collaborate with humans, adapt to changes in workflows, and learn new skills without reprogramming.”

The result will be advanced robots, including existing robots enhanced by AI at the edge, as well as new AI-enabled models. Eventually, humanoid robots will take their place in automotive factories. Humanoid robots are attractive because they offer greater flexibility than traditional ones, and are suited to operating in environments designed for humans, using the same physical tools.

Advanced robots like this bring multiple benefits, summarized in the table below.

Robots’ flexibility will help businesses respond in an agile manner to an unpredictable environment. For example, switching a production line from one model to another should happen rapidly and with minimal human intervention.

Equally importantly, automation through advanced robotics will enable human experts to focus on innovation rather than running plants, enabling automotive companies to maintain their market lead despite mounting competition from digital-native newcomers. At the same time, offering this more cognitively stimulating work will help the companies to recruit and retain scarce talent.

Advanced robots like this bring multiple benefits, summarized in the table below.

Robots’ flexibility will help businesses respond in an agile manner to an unpredictable environment. For example, switching a production line from one model to another should happen rapidly and with minimal human intervention.

Equally importantly, automation through advanced robotics will enable human experts to focus on innovation rather than running plants, enabling automotive companies to maintain their market lead despite mounting competition from digital-native newcomers. At the same time, offering this more cognitively stimulating work will help the companies to recruit and retain scarce talent.

Challenges of humanoid robots

It’s worth noting here that advanced humanoid robots in automotive factories face challenges such as:

• High cost: The initial investment and ongoing maintenance for robots can be extremely expensive.
• Battery limitations: Battery life is a significant bottleneck, with current models often limited to just a few hours of active work per charge, so that frequent swaps are needed to maintain continuous operation.
• Speed and strength limitations: Many advanced robots are currently slower and less powerful than specialized, fixed industrial automation systems or human workers.
• Safety concerns: While robots are designed to be collaborative, ensuring absolute safety in fast-paced environments still requires extensive testing, and a loss of power could still pose a risk.
• Programming complexity: Effectively implementing and maintaining robots requires advanced skills and configurations.

Taking account of these challenges, the new robots are best suited to complex multi-purpose tasks where their ability to work in human-centric environments and collaborate safely is a key advantage.

Seize the opportunities of advanced robotics

So there is a strong business case for automotive companies to adopt the next generation of industrial robots. To do so successfully, several preparatory actions are needed.

Understand the opportunities and harness innovative technologies and techniques

A major area of opportunity for automotive companies is opened up by cobots – AI-powered robotic systems that handle repetitive, high-precision, and strenuous tasks alongside human workers. We discussed this topic earlier in this series. Examples of cobot applications in automotive include:

• Materials handling: Cobots can work alongside humans, carrying out assembly line tasks like materials handling with precision and speed.
• Adaptive assembly: Instead of following fixed programs, AI-enabled robots can adapt their movements based on real-time visual feedback, allowing them to handle the variability of complex tasks like cable routing or fitting transmission components.

Cobots are one of several areas of opportunity for automakers to explore. But to realize these opportunities, auto manufacturers need to get to grips with a range of new concepts – including advances in AI and related technologies, novel applications of those technologies, and better ways of working with them. Here are just a few examples:

• Hybrid AI combines generative AI with other models based on a new type of AI: liquid neural networks (LNNs). Compared with large language models, these are easy to train, use few computing resources, and produce results that are both accurate and explainable.
• As robots take on more and more work within the factory, better ways of assuring reliability become vital. For example, digital twins will increasingly be used to validate robotics solutions – and other aspects of an automated production line – prior to deployment. The explainability of LNNs can also facilitate validation.
• Sustainability should be engineered into robotics solutions from the outset – and sustainable energy use is a major consideration given the power-hungry nature of many AI models. As Daniela Rus says in her conversation with Capgemini: “One key strategy is to develop more efficient AI architectures. For example, LNNs offer strong performance with fewer parameters and lower compute needs.”

“A new generation of multi-purpose, AI-powered robots is round the corner: one that will offer major advantages to companies implementing intelligent manufacturing strategies. As experienced robotics users, automotive companies are strongly positioned to leverage these new robots, taking advantage of their flexibility, collaborative capabilities, and ability to respond to their environments. Apart from the challenge posed by the scarcity of the specialist skills required for implementation, a sound implementation strategy will be crucial to ensure acceptance, learning to collaborate with physical AI instead of competing with it.”

Nicolas Rousseau, EVP, Chief Digital Engineering & Manufacturing Officer, Capgemini Engineering

Manage technical and organizational transformation

Although automotive companies are already expert in many aspects of robotics, they will need to adapt to the new generation of robots. Technical tasks will include:

• Dealing with the diversity of hardware ecosystems
• Integrating data into diverse formats
• Developing the complex AI algorithms needed to control operations
• Achieving the low latency required for mission-critical processes

The human side of the organization will also need to adapt. An earlier article in this blog series emphasized that the true power of innovative technology in manufacturing lies in using it to augment and complement human capabilities.

To make this relationship work in practice, careful thought is needed. For example, the organization must systematically determine which tasks can be safely assigned to robots and which should continue to be controlled by humans. Employees will need to be upskilled to work with robotic systems.

Most crucial and most challenging of all, it will be vital to secure the workforce’s trust in the technologies they are expected to work with. Implementing more transparent technologies such as LNNs has a vital part to play here, along with realism about what robots can and cannot do – and honest discussion about the impact of automation on employment prospects.

Collaborate with ecosystem partners to secure the skills and assets needed

In such a fast-moving field, building relationships with leading players is often the most reliable and cost-effective way to ensure ongoing access to the latest thinking and tools.

Complementing a strong team of in-house robotics experts, our own Capgemini AI Robotics and Experiences Lab works with partners such as Nvidia, Unity, Dassault Systèmes, Siemens, Microsoft, Google Cloud, and AWS. We also partner with specialist innovators such as Liquid AI.

This range of internal and external capabilities enables us to offer our clients roadmaps and prototypes that are both innovative and realistic.

Capgemini, robotics, and intelligent manufacturing for automotive

Robotics is an integral part of Intelligent Manufacturing Services for Automotive by Capgemini. Our automotive manufacturing team collaborates closely with our specialist robotics lab, and draws on knowledge and expertise from across our multi-industry practice. Here’s one of our latest announcements in this area.

The table below summarizes our approach to supporting clients’ journeys from industrial automation to physical AI.

Please contact us to find out how we can help your automotive company leverage advanced AI-enabled robotics technology and techniques across its manufacturing operations.