Hitachi and FANUC Team Up on Physical AI for the Factory Floor

Hitachi is collaborating with FANUC to integrate AI with robotics, targeting what the companies describe as the leading edge of automation research.

Why Physical AI Matters

The collaboration focuses on physical AI — the intersection of robotics and artificial intelligence — which represents the forefront of automation research.

"The marriage of robots with artificial intelligence is the core of physical AI, and it's the leading edge of automation research."

For manufacturing professionals, this signals a shift beyond traditional programmed automation toward systems that can adapt and learn on the production line.

Combining Complementary Strengths

The partnership pairs each company's core expertise rather than duplicating capabilities.

"The partnership combines Hitachi's HMAX Industry AI expertise with FANUC's AI enabled industrial robots and control technology."

Hitachi brings its HMAX Industry AI platform to the table, while FANUC contributes its AI-enabled industrial robots and control technology — a division of labor that plays to each firm's established strengths.

Hitachi Factories as "Customer Zero"

Rather than launching with external customers, the plan starts in-house. Hitachi group factories will serve as the proving ground where AI learns real production activities.

"The plan will use Hitachi group factories as 'customer zero', where AI will learn production activities."

This approach lets the partners refine the technology under actual production conditions before broader deployment — a pragmatic strategy that acknowledges how difficult adaptive automation is to get right.

Target Applications

The initial use cases address two persistent pain points in flexible manufacturing:

"Specific applications will include picking parts of various shapes and performing changeovers when production items change."

Parts picking of varying geometries and rapid changeovers are exactly the tasks where conventional automation tends to struggle. If the collaboration delivers, these applications could meaningfully improve flexibility on mixed-model production lines.