How does the platform decide what a robot may do?

It compares user attribute information such as age against rules registered for each usage setting, and stops any action not permitted before sending only approved instructions to the robot.

What standard does it use to connect AI agents?

It uses A2A, a standard protocol for AI agents to communicate. Permission control, action-history tracking and situation-aware behavior generation were developed independently by Docomo.

What restaurant use cases does Docomo envision?

It anticipates choosing a serving robot for orders placed via an AI agent and preventing alcohol service to minors or delivery of allergen-containing food, with robots also stepping aside when people approach.

Large Language ModelsRoboticsAI Safety & AlignmentImpress Watch5m agoOct 27, 2026, 19:01 JST

NTT Docomo links AI agents to robots with Physical AI platform

Send on LINE

3 Key Points

What happened
NTT Docomo developed "Physical AI連携基盤," which unifies linking AI agents to robots, controlling permitted actions, and situation-aware behavior. It uses the A2A protocol, with permission control and action logging built in-house.
Why it matters
Permissions are checked against user attributes such as age and per-setting rules, so disallowed actions (serving alcohol to minors, allergen dishes) are blocked before reaching the robot — a safety layer for home and restaurant use, it appears.

WHO IT HITSRestaurant operators and robot makers targeting food-service deployment are the clearest near-term audience, since the body cites serving robots that avoid alcohol for minors and allergen dishes; home users also stand to benefit from the safety controls.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, every morning.

Get these free every morning

Context & Analysis

The platform rests on three functions Docomo bundles into one base: linking AI agents to robots, controlling whether an action is allowed, and generating considerate behavior for the surroundings. The body stresses that the whole chain from a user's instruction to the robot's motion is managed and traceable, and that the platform keeps attribute data internally, sending the robot only motion instructions.

A2A, a standard protocol for agent-to-agent communication, provides the connection layer, while permission control, action logging and context-aware behavior were built in-house. That split suggests Docomo aims to sit between personal agents and many robot makers rather than build robots itself, a reading consistent with its stated plans for joint research with robot manufacturers and proof-of-concept work with businesses.

FAQ

How does the platform decide what a robot may do?: It compares user attribute information such as age against rules registered for each usage setting, and stops any action not permitted before sending only approved instructions to the robot.
What standard does it use to connect AI agents?: It uses A2A, a standard protocol for AI agents to communicate. Permission control, action-history tracking and situation-aware behavior generation were developed independently by Docomo.
What restaurant use cases does Docomo envision?: It anticipates choosing a serving robot for orders placed via an AI agent and preventing alcohol service to minors or delivery of allergen-containing food, with robots also stepping aside when people approach.

Impress WatchRead Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.