Connecting Industrial Robots with Real-World Production Environments

Artificial intelligence is increasingly moving beyond software applications and into machines that interact with the physical world. In manufacturing, this development is often described as Physical AI: intelligent systems that perceive their surroundings, make decisions, and perform physical tasks.

Aptiv announced two separate strategic collaboration agreements with robotics companies TARS and Shanghai Electric on November 6, 2026. The agreements focus on deploying robots in manufacturing operations and integrating Aptiv technologies into new robotic platforms.

According to Aptiv, hundreds of robots from the two partners are being gradually deployed on its production lines. Initial activities are moving from validation toward broader deployment.

The announcement provides an example of how automotive manufacturing environments can become real-world testing grounds for industrial robotics and Physical AI.

Why Manufacturing Is a Critical Environment for Physical AI

A factory is a demanding environment for intelligent machines.

Robots must operate around equipment, materials, people, and changing production requirements. They need to recognise objects, understand task instructions, coordinate movements, and respond to unexpected conditions.

A system that performs well in a controlled demonstration may behave differently when exposed to real production variability.

Manufacturing therefore provides an important environment for evaluating the reliability of Physical AI.

Production data can help developers identify limitations in perception, decision-making, motion control, and task execution. The information can then be used to refine robotic systems before wider deployment.

Aptiv’s Role: Computing and Intelligent Power Distribution

Aptiv is contributing technologies that support the computing and electrical architecture of robotic platforms.

The company says its high-performance computing platforms and intelligent power distribution solutions will be among the first technologies integrated into robots from TARS and Shanghai Electric.

These systems address two important requirements.

High-performance computing: Robots need computing resources to process sensor information, run perception algorithms, make decisions, and coordinate motion.

Intelligent power distribution: Robotic systems contain multiple electrical loads, sensors, actuators, and computing modules. Power must be distributed reliably while supporting the operational requirements of the platform.

Aptiv says initial production involving these integrated technologies is planned by the end of 2026.

From Pilot Deployment to Manufacturing Scale

One of the most important aspects of the announcement is the use of Aptiv’s own manufacturing facilities.

Deploying robots in production allows the companies to evaluate how systems behave during actual industrial operations.

The collaboration is expected to generate information about:

• Robot performance under real manufacturing conditions.
• Reliability during repeated industrial tasks.
• Perception and decision-making in changing environments.
• Integration with existing factory operations.
• Power and computing requirements.
• Opportunities to improve production efficiency.

Aptiv describes this operational feedback as an important part of developing and commercialising Physical AI technologies.

However, gradual deployment should not be interpreted as proof that all manufacturing tasks have become fully autonomous. The specific tasks, deployment locations, and production responsibilities will determine the practical impact.

Engineering Challenges in Industrial Robotics

Integrating robots into an existing factory requires more than installing a robotic platform.

Mechanical engineers must consider structural strength, actuator loads, payload capacity, thermal behaviour, and serviceability.

Electrical engineers must address power distribution, wiring architecture, communication, electromagnetic compatibility, and functional safety.

Control engineers must ensure that robots can coordinate their actions with production equipment and respond safely to changing conditions.

AI engineers must develop perception and decision-making systems that remain reliable when objects, lighting, layouts, and production conditions vary.

These disciplines must work together to deliver a system that is not only intelligent but also predictable and maintainable.

The CAD, CAE, and Digital Manufacturing Connection

CAD supports the design of robotic structures, mounting interfaces, housings, and mechanical components.

CAE can be used to evaluate structural loads, vibration, thermal performance, and the mechanical behaviour of robot assemblies.

Digital manufacturing tools help engineers study how robots interact with production lines, workstations, material flows, and existing automation systems.

In more advanced environments, digital twins can provide a virtual representation of manufacturing operations, allowing engineers to evaluate layout changes and automation strategies before implementing them physically.

Why This Development Matters to the Automotive Industry

Automotive manufacturing has long been a major driver of industrial automation. However, traditional automation is often based on highly structured tasks and carefully designed production cells.

Physical AI aims to extend automation into tasks that are less predictable and require greater adaptability.

For automotive manufacturers, this could create opportunities in material handling, internal logistics, inspection, machine tending, and other repetitive operations.

The long-term value will depend on whether these systems can achieve the required levels of safety, reliability, productivity, and economic performance.

Aptiv’s partnerships illustrate an approach in which robotics technology is developed and refined inside real manufacturing environments rather than relying exclusively on laboratory demonstrations.

Source: Aptiv – Accelerates Robotics Expansion With Industry Partners

Related technology: Physical AI, industrial robotics, high-performance computing, intelligent power distribution, smart manufacturing.