On October 21, Mitsubishi Electric announced that it has developed
a machining system in which humanoid robots and CNC machine tools
work together through physical AI.
When a person gives instructions in natural language,
the AI builds a work plan and assigns tasks to both.
The company aims for practical use in 2030,
toward automating tasks such as setup and cleaning.
Background: Human-Oriented Tasks and the Effort of Connecting Equipment
The company explains that, amid labor shortages at machining sites,
demand is growing for automating tasks handled by operators.
However, replacing tasks that assume human physical characteristics,
such as setup and cleaning, with industrial robots
requires capital investment in production lines tailored to the robots.
Even when using humanoid robots,
connecting them to machine tools so that each can understand the other’s status
and work together has required developing communication functions
and control sequences for each robot model.
The new system addresses this coordination between machines.
The company describes it as an industry first
as an NC system designed for humanoid robot collaboration
(as of October 21, 2026, according to Mitsubishi Electric’s research).
How It Works: AI Combines Registered Tasks and Manages Execution
The system consists of three components:
an AI machining task planner, a CNC machine tool equipped with physical AI,
and a general-purpose humanoid robot.
Physical AI is AI technology that controls and optimizes
physical movements in the real world.
The planner and the CNC are equipped with physical AI newly developed
using Maisart, the company’s AI technology brand.
The planner analyzes natural language instructions
and combines the necessary tasks from a list of pre-registered work tasks.
Following that plan, it assigns work to the machine tool and the robot
and manages the machining operation.
The two exchange operating information in natural language to work together.
What Changes: Reducing Development Effort at Deployment and Manual Work
The company says the system can work flexibly
regardless of the robot model,
and by reducing advanced system integration work,
can significantly shorten the time to deployment.
The tasks targeted for automation include setup, cleaning, tidying up,
removing finished products, transport, inspection, and troubleshooting.
The announcement does not give specific figures
for the reduction in deployment time or labor savings.
LC Signal’s Perspective: Connecting the Work That Remains Around Machine Tools
The scope extends beyond machining itself
to preparation, cleaning, inspection, and more.
The approach of combining each of these as registered tasks
is interesting for seeing how much of the series of tasks
people have handled can be entrusted to machines.
The company aims to deepen coordination between machines
based on physical AI and to realize autonomous, unmanned machining factories.
Toward practical use, the company will advance its functions
and conduct demonstration tests,
and plans to exhibit a demonstration at JIMTOF2026,
which opens on October 26 at Tokyo Big Sight.
We will also watch how the system actually works,
from planning tasks to coordinating the machines.
Key Sources
-
Mitsubishi Electric official announcement (Japanese), October 21, 2026:
業界初、ヒューマノイドとCNC工作機械が連携する加工システムを開発
(Industry first: Machining system in which humanoid robots
and CNC machine tools work together developed) -
Mitsubishi Electric official announcement, full text PDF (Japanese):
業界初、ヒューマノイドとCNC工作機械が連携する加工システムを開発（全文）
(Industry first: Machining system in which humanoid robots
and CNC machine tools work together developed, full text)
This article was prepared with AI assistance using publicly available information
as of October 22, 2026.
Announcement dates follow the dates shown on the official pages.
The developed system is distinguished from the plan for practical use targeted for 2030.
“LC Signal’s Perspective” reflects this publication’s interpretation
of the available information.