Universal Robots Launches Gen 7 Platform with Edge AI Capabilities for Plastics Manufacturing

Universal Robots’ new Gen 7 platform provides the physical AI infrastructure for robotics applications within plastics manufacturing.Universal Robots

Artificial intelligence is transforming plastics manufacturing, as robotics suppliers and equipment makers deploy machine learning and advanced data analytics to build more agile, efficient, and sustainable production lines.

By enabling improvements from smarter material handling to dynamic production scheduling that optimizes supply chain readiness, AI is taking plastics manufacturing to the next level of efficiency.

Plastics equipment suppliers like KraussMaffei and robotics experts like Fanuc and Universal Robots are deploying AI to help manufacturers better understand and optimize their operations. AI-driven robotics from Universal Robots (UR), for example, are achieving multiple production savings, including:

• 40% faster cycle times with UR's OptiMove motion control software, which optimizes trajectory paths better than classically written robot programs.

• 3D part localization on the fly in 0.5 seconds when combined with Inbolt's robot intelligence platform, compared with five- to 10-second delays when waiting for mechanical shakers, pneumatic clamps, or indexing tables to physically align parts.

Related:Nanotec Unveils Compact LBA80 Linear Actuator

AI advances robot performance

Industrial robot suppliers are introducing AI-powered capabilities into their platforms and controllers to optimize processes before production even begins. AI-enabled robotics give plastics manufacturers greater visibility into how production systems perform and how process decisions affect operations.

At Fanuc, "our Physical AI strategy combines robotics, simulation, digital twins, and AI-powered technologies to help manufacturers evaluate production scenarios, identify inefficiencies, and validate workflows before they are implemented on the factory floor," said Amar Dhaliwal, general manager of FANUC America's Automation Systems Group. "By creating and testing production processes in a virtual environment, manufacturers can make more informed decisions that improve overall operational efficiency and reduce costly trial and error during deployment."

Fanuc 's Physical AI strategy lets robots "perceive their environment, make decisions, and take action in ways that were not possible with traditional fixed automation," Dhaliwal explained. "Fanuc combines real-time robot control, AI-driven perception, simulation, and digital twins to help robots operate more effectively in dynamic manufacturing environments. AI-enabled systems can use visual and sensor data to understand what is happening around them and adapt to changing conditions."

Not only is AI enabling robots to work better, but it's making automation "more approachable" for human operators, he continued.

Related:Plastics and Rubber Robot Orders Surge

"Technologies such as simulation, digital twins, AI-assisted programming, and natural-language commands can reduce complexity and enable more employees to interact with automation systems. Rather than replacing people, these tools help workers become more productive by allowing them to focus on higher-value activities such as process improvement, troubleshooting, and system optimization."

Meanwhile, Universal Robots has just launched its new Gen 7 platform, which provides the physical AI infrastructure for robotics applications within plastics manufacturing.

UR's new CB7 Core controller delivers 40% more computing power in a 30% smaller footprint, running the open PolyScope X operating software. This allows third-party AI vision and sensor algorithms to run locally at the edge, directly on the robot controller, without requiring host PCs or cloud connections, explained Christopher Buse, UR application engineer.

"Plastics manufacturing, such as injection molding, thermoforming, and blow molding, frequently deals with material variability caused by thermal shrinkage, part warping, shifted dunnage, or uneven sprue trimming," Buse noted. "Traditional automation relies on rigid mechanical jigs to hold parts in fixed locations; when parts vary, misloads and mold crashes create scrap and downtime."

Related:Why Automation Demand Is Rising as Factories Downsize

The new tool flange on Gen 7 robots "routes power, high-speed Ethernet, and safety signals directly through the interior of the robot joints. This eliminates external cabling harnesses that traditionally suffer from bend-radius limits, cable fatigue, or snagging in tight molding environments."

UR partners including Inbolt and Cognex have developed solutions that run on the Gen 7 platform. These solutions include:

• 3D vision guidance and part localization: Inbolt provides the Inbolt Standard 3D vision guidance system. Using a wrist-mounted 3D camera, Inbolt compares live parts to their 3D CAD models, locating parts across six degrees of freedom (6DoF) in as little as 0.5 seconds. By enabling the cobot to adapt to warped or misaligned molded parts automatically, Inbolt eliminates rigid mechanical fixtures.

• Quality inspection and defect detection: Cognex supplies 2D and 3D vision systems powered by edge learning tools. Operating natively via PolyScope X, Cognex inspects molded components immediately upon extraction, performing optical character recognition, verifying component placement, and identifying surface defects on the fly. Detecting non-conforming parts at the mold prevents defective inventory from moving downstream, cutting material waste.

In each case, the 3D camera (Inbolt) or vision camera (Cognex) mounts directly to the robot wrist and plugs straight into the internal flange Ethernet port.

"The [Inbolt] 3D camera captures a snapshot of the scene," Buse explained. "Inbolt's AI software, running at the edge on the CB7 controller, compares the image against the part's 3D CAD digital twin and calculates the exact spatial offset across X, Y, and Z coordinates, plus roll, pitch, and yaw angles."

To achieve mid-air trajectory adjustment, he added, Inbolt "passes the calculated 6DoF offset vector directly into the UR program tree as a native script command."

Overall, built-in force-torque sensing inside UR's Gen 7 tool flange gives the cobot "a digital sense of touch," Buse said. "During delicate part insertion, machine tending, or sprue trimming, the robot senses insertion forces in real time and automatically adjusts force levels to avoid cracking fragile plastic parts or damaging tooling."

AI at machine level

Employing AI beyond just the robotics stage — across the entire plastics production automation ecosystem — can leverage process and material data to stabilize production, improve material utilization, and reduce waste.

KraussMaffei, for instance, embeds intelligence directly into the production environment to make advanced technology practical for everyday manufacturing, said Markus Bauer, managing director of KraussMaffei Technologies.

"KraussMaffei currently focuses AI and intelligent algorithms primarily on optimizing the manufacturing process and supporting production teams," Bauer explained. "With socialProduction, we have a tool that translates large, unmanageable amounts of data (data lakes) into clear actions and instructions."

KraussMaffei is developing an AI-powered assistant called "K-AI" within the socialProduction environment to give manufacturing teams faster access to relevant information and support data-driven decision-making, he added. "Our focus is on applying AI and intelligent algorithms where KraussMaffei has deep process expertise: directly at the machine and at production-process level."

To improve recycling processes, reduce scrap rates, and enable circular-economy initiatives, KraussMaffei is taking quality control to the next level with its Autonomous Quality Control system. The system measures actual part weight after every cycle and feeds the result directly back into the machine control system.

"Variations caused, for example, by changing material densities in recycled materials, additives, or fillers can be detected and compensated for automatically," Bauer said. "This helps to reduce scrap, stabilize the process, and keep material consumption within tight tolerances."

Circularity extends beyond injection molding, he added.

"KraussMaffei Extrusion provides technologies for mechanical recycling, advanced chemical recycling, and PUR recycling, including the ZE BluePower twin-screw extruder for processing and upgrading recycled material streams. In this context, intelligent control does not replace recycling technology. It helps manufacturers manage the variability of recycled materials and use them more reliably in industrial production."

To help customers overcome the challenges of integrating AI technology in the plastics manufacturing environment, KraussMaffei's integrated technology initiates automatic corrections instead of just generating additional data for operators to interpret. Solutions such as APCplus and Autonomous Quality Control are integrated into the machine and its control architecture.

"APCplus compensates for variations in the molding process, while Autonomous Quality Control creates a closed control loop between actual part quality and the machine process," Bauer said.

At the same time, Bauer noted socialProduction addresses another challenge of digital manufacturing: turning growing amounts of production data into information that is useful for people on the shop floor.

"In manufacturing, AI delivers value when it is connected to reliable process data and a detailed understanding of the underlying production process,” Bauer concluded.