The automotive industry, like all industries, is going through a major transformative change, the biggest it’s seen in nearly a hundred years. The major drivers that are present today have been there since the inception of robotics in the manufacturing process as far back as 1961 when they were first used in an industrial way. Efficiency and accuracy, labour reduction and of course cost savings were the beckoning promise of robotics and automation. And now with Industry 4.0—with the integration of technologies like the Internet of Things (IoT), cloud computing and big data—automation now touches every aspect of the value chain, from suppliers to end customers. As a result, the automotive industry is more connected and adaptable than ever before.
In the face of adversity
As everyone knows, the automotive sector has been experiencing a global shift, a set of new challenges as the world races to find new environmental and reliable energy sources that are not hydrocarbon based. Add to this a Chinese influx of EV competitors putting an enormous stress on European and US brands to be more competitive and cost efficient, especially in the production process. The threat of tariffs being imposed by the US and other governments also looms large. "There are so many crises, a whole world of crises. When one crisis is over, another is coming up," is how Simon Shütz, a spokesman for the German Automotive Industry Federation (VDA), put it to the BBC.
Part of the answer is automation: autonomous mobile robots being guided by a central command ala Automated Guided Vehicles (AGV) or the new onboard philosophy of Automated Mobile Robots (AMR) whereby they use machine learning and other tools to make decisions. According to GlobeNewswire, the global industrial robotics market was valued at around $48.5 billion in 2022 and is expected to reach approximately $142.8 billion by 2032. The automotive industry recently hit a milestone of one million robot units in operational stock according to the International Federation of Robotics (IFR) becoming the world’s most automated industry, outpacing even the electronics industry. “The automotive industry effectively invented automated manufacturing,” notes Marina Bill, the President of IFR. Naturally, all of this is driven by cost pressures, especially in terms of labour. By using automation, Tier 1 European and US OEMs are regaining their competitive advantage that Chinese automotives have threatened in the last few years, especially with the critical race to make EVs price competitive and range sufficient (the €25K/500 km range delta). By retooling with new automation driven methodologies like the Tesla unboxed construction, the automation race is on. Tesla, a pioneer regarding factory automation and robots, has said that introducing more automated equipment at Tesla as part of a goal to cut the costs of making future models by 50%, according to CBT News. The benefits are glaring: cost decreases by up to 50% driven by reduced capital investment in fixed conveyor systems; more efficient resource utilisation and streamlined logistics; factories seeing a 40% reduction in real estate and energy footprint and enabled urban manufacturing; parallel processing that allows different vehicle sections to be assembled simultaneously which in turn shortens production times and increases output capacity. And importantly, increased flexibility as the modular nature of unboxed assembly should allow for easier model changes and more efficient production of multiple models on the same line. This scale up/scale out will be a highly attractive feature of the new production lines that will come up under the Unboxed methodology and will need automated vehicle movers to replace in-line and end-of-line moving. This phenomenon can be called Flexiline as automation replaces the static production line of the Henry Ford era that has held sway for the last 100 years. In the Flexiline production scenario, automation enabled vehicle movers will move the vehicle through the different stations to larger, more static robotics (window shield installation, interior insertion etc) and even out into peripheral areas like vehicle storage staging or custom accessorising bays. As it’s scalable and configurable, these mobile robots will be a much greater part of the production line than they are today.
Many fleet footed followers are taking the unboxed philosophy onboard, especially Asian manufactures. But slowly, European brands are taking up the challenge to stay relevant and innovative. BMW's iFactory concept envisions the future of its manufacturing plants with a strong emphasis on flexibility and modularity. The iFactory will integrate unboxed assembly principles, allowing BMW to adapt production lines to new models and technologies quickly. In the coming year, more Tier 1 OEMs and smaller volume players will use Unboxed, with its critical reliance on automation, to radially reshape the automotive production line.
The Unboxed methodology, in a nutshell
- Modular Construction: Vehicles are assembled from large, pre-manufactured modules.
- Parallel Processing: Multiple vehicle sections are constructed simultaneously, not sequentially.
- Flexible Assembly Sequences: The process can adapt to different models and configurations.
- Automated Guided Vehicles (AGVs): These transport modules within the factory, replacing fixed conveyor belts.
Source: Forbes
Mobile robots, AI, ML, DL and swarm: the next steps in automotive automation
While the push to automate more and more for all the reasons listed above is clear, how that’s being implemented is a lesson in disruption and innovation. The first AGV was introduced in 1953 by Barrett Electronics of Northbrook, Illinois, a tow truck that followed a wire in the floor instead of a rail. Though there are significant advancements in AGV from laser guided systems to AI integrations, the basic idea of AGVs stays the same namely that they have a predetermined task route to complete the intended command. They can do this planned route repetitively and with high efficiency. One should think of AGVs as a train: consistent, very linear and very effective at consistently structured tasks. This has been the dominant technology in automation.
Lately, however a disrupter has come to into the AGV space and is forcing a synthesis. Called Automated Mobile Robots (AMRs), these were developed as far back as the 1950s (the first pair were called Elsa and Elmer) but it was really in the hospital industry in the 1990s as an automated solution for delivering meals etc in hospital settings that their influence spread. They are different to AGVs in that they don’t rely on a pre-determined path or task route to accomplish their goal. Instead, they use onboard sensors and computing to process data and make independent decisions to accomplish a task. In fact, with onboard intelligence and decision making enabled, AMRs are can make independent decisions that the original task designer couldn’t have thought of. Instead of the route being from A-B-C the AMR realizes that B-C-A is more efficient and can demonstrate that with its machine learning analytics. As AGVs are to trains, think of AMRs as cars, capable of lateral decisions and more real time analysis and decisions to accomplish its goals.
And it’s important to remember that in both cases—AGV and AMR—the analogy of trains and cars is a fitting one. Both as trains and cars exist in our world right now for different appropriate use cases, so too AGVs and AMRs have a similar situation happening to them. So much so that a synthesis is happening whereby a new term, Mobile Robots, is aggregating the two approaches into a more meta term. “The strict categorization of mobile robots into categories like “AGV” or “AMR” is no longer up to date,” explains Mathias Behounek, CEO of SAFELOG, a mobile robotics operator in Germany. “Our new names are intended to show that SAFELOG robots no longer need rigid definitions. They are as flexible and versatile as their tasks,” continues Behounek.
The next steps for automotive automation are happening quickly. Artificial Intelligence (AI) machine learning (ML) and Deep Learning (DL) algorithms are predictably taking the place of labour to do tasks in the automotive environment as IoT connected driven mobile robots through 5G or wifi networks are enabled. While that connectivity allows cloud based computation in real time, it also allows the next level of mobile robotic: swarm networking. In swarm networks, mobile robots share data from their place in the operating space with all other mobile robots on their network, allowing AI/ML/DL decisions to be made to perform the tasks optimally on every mobile robot and collectively. For instance, let’s say that 20 mobile robots are operating in an automotive production line in a flexiline scenario with swarm connectivity. One of the mobile robots breaks down and can’t complete its task. Instead of a human manager making a decision on how to fix this problem, the connected mobile robots will crowd source all their data from each other to figure out the optimal solution for all the tasks that are being performed.
Another scenario is a safety one whereby swarmed mobile robots are sharing information on what’s happening in an environment. Mobile Robot number 1 shares that it sees an employee moving in a direction of travel that will intersect with Mobile Robot number 4 which is coming up an aisle nearby. By swarming this information, an accident is averted that would’ve happened were Mobile Robot number 4 not connected to the swarm network.
Taken together, the automation trend will use all of these technological solutions to deliver on the needs of the automotive industry as it goes through a period of challenges. But just like innovation periods previously, those needs will be addressed and the mobility space will transform into new, more vital configurations.
