LG CNS launches platform to manage mixed robot fleets

LG CNS, the IT services arm of South Korea’s LG Group, launched its PhysicalWorks platform to train and manage mixed robot fleets through one software layer.

At a demo, four robots from Unitree, Deep Robotics, Dexmate, and Bear Robotics moved boxes without remote control, and one handoff over two to three meters took 90 seconds.

LG CNS said the system combines simulation and video training with software that reassigns work in real time.

It switched in a Bear Robotics cart after a staged emergency diverted a quadruped to patrol duty.

The company said it is running over 20 proof-of-concept projects, while an executive said revenue may take one to two years.

🔗 Source: The Korea Herald

🧠 Food for thought

Implications, context, and why it matters.

LG’s robot platform draws on four decades of factory software experience

The move grows out of LG CNS’s work as a systems integrator, a company that connects and runs complex business software systems, rather than a sudden shift into hardware 1.
For 40 years, the company has built IT backbones for manufacturers. That background includes linking older production software, which it treats as an advantage 1.
PhysicalWorks builds on LG CNS tools such as Real Time Dispatcher (RTD), which sets task priorities and logistics movement conditions in real time. It can also help control logistics equipment including Automated Guided Vehicles (AGVs), which are driverless vehicles used to move materials in factories 2.
The launch came after 11 months of preparation. That period included an investment in Skild AI, a US startup building AI systems for robots, plus a stake in robotics firm Dexmate 1.

One control layer for mixed robot fleets

The platform manages robots from different manufacturers through one control layer. That addresses a fragmented market where machines from separate vendors often need custom engineering to work together 1.
The setup could make automation easier to adopt, letting businesses pick the best robot for each job without being tied to one vendor ecosystem 3.
The system can cut robot deployment from several months to about one or two months 4.
Through its partnership with Skild AI, the platform adds a Robot Foundation Model (RFM). The model aims to make robots more adaptable by letting them learn from workplace photos and video data, then act on their own instead of requiring task-specific development for each action plus direct control at every step 5.