Arm Launches Robotics Capability Framework For Intelligent Robots

(MENAFN- Robotics & Automation News) Semiconductor and software design company Arm has introduced a Robotics Capability Framework designed to give the robotics industry a common language for describing and comparing the capabilities of increasingly intelligent machines.

The chip architecture company says the framework is intended to address growing ambiguity around terms such as“autonomous”,“intelligent”,“adaptive”,“collaborative” and“self-improving” as artificial intelligence becomes more deeply integrated into robots.

Arm describes the initiative as an architecture-agnostic reference framework covering robotic capability, operating context, supervision and assurance. Rather than specifying how robots should be designed, the company wants to establish a consistent way of describing what systems can actually do and under what conditions.

Richard Grisenthwaite, chief architect at Arm, says:“At Arm, we believe the next phase of robotics requires more than just advances in AI and compute. It needs a shared foundation that allows intelligent robotic systems to be described consistently without constraining innovation.”

Arm argues that inconsistent terminology is already making robotic systems more difficult for customers to compare and integrate, while also creating problems for developers, regulators, insurers and other organizations attempting to assess their capabilities and associated risks.

The issue is becoming more significant as robots move beyond tightly controlled industrial environments and into warehouses, hospitals, agriculture, infrastructure inspection and other settings where they operate alongside people and must respond to changing conditions.

Arm says more than 4.28 million industrial robots are already operating in factories worldwide, citing International Federation of Robotics figures, while advances in AI, sensing, compute and software are expanding the range and sophistication of tasks robotic systems can perform.

From 'reactive' to self-improving robots

The proposed Robotics Capability Framework combines two complementary ways of describing a robot.

The first is a series of levels describing progression from reactive behavior towards“deliberative, adaptive, contextual, cognitive and self-improving intelligence”.

The second describes the areas in which those capabilities are expressed, including perception, localization, motion, manipulation, decision-making, learning, interaction, safety, security, lifecycle management, interoperability and ecosystem integration.

Arm says the combination should make it possible to create a capability profile for a robot rather than attempting to characterize its intelligence with a single term.

This could help distinguish, for example, between a fixed industrial robot designed to operate predictably within a tightly controlled environment and a mobile manipulator working near people, which may require greater contextual awareness, interaction capabilities and fallback behavior.

Similarly, fleets deployed across multiple sites could require more sophisticated lifecycle management, update governance and shared learning capabilities.

Arm emphasizes that the framework is not intended to establish a particular hardware architecture, AI model, software stack or cloud-edge configuration.

Instead, it is designed to provide what the company describes as a stable reference layer between broad claims about robot intelligence and the detailed technical choices used to implement those capabilities.

The company says this could ultimately make procurement and integration easier while giving regulators and insurers more neutral terminology for evaluating safety, assurance and risk.

Industry-wide initiative

Arm is positioning the Robotics Capability Framework as a starting point rather than a finished standard or an Arm-only initiative.

The company is inviting robot manufacturers, customers, integrators, silicon and software companies, regulators, insurers, researchers, standards organizations and other industry participants to contribute to its development.

The manifesto lists supporters including Bosch, Infineon, Lenovo, McKinsey & Company, QNX, Psyonic, RobotecAI and StradVision, among others.

Arm says the framework should follow several principles: capability descriptions should be clear and understandable; the framework should describe capabilities rather than prescribe implementations; vendors should remain free to differentiate their products; and the framework should accommodate the wide variety of machines and operating environments found across robotics.

Grisenthwaite says:“We see Robotics Capability Framework as a practical starting point, and invite robotics companies, developers, customers, regulators, researchers and industry partners to help shape it into a trusted foundation for the next generation of intelligent robotics.”

Arm expands physical AI initiative

The Robotics Capability Framework forms part of a wider expansion of Arm's activities in physical AI.

Arm says it is extending its Arm Total Design program into physical AI, bringing together more than 80 companies including AWS, Hugging Face, Liquid AI, QNX and Unitree Robotics. The initial focus includes establishing the Robotics Capability Framework as a common way to“describe, compare and communicate” the capabilities of intelligent autonomous machines.

The physical AI initiative was one of several announcements presented by Arm as part of its broader strategy for what it calls the emerging“agentic” computing era.

At the edge, Arm introduced CSS for Mobile 2, incorporating its new C2 Ultra CPU and Mali G2-Ultra NX GPU. In cloud infrastructure, the company announced Neoverse CSS N4 and highlighted growing momentum around its AGI CPU platform.

Arm also introduced the Arm AI Portal, intended to help developers and AI agents find optimized models, performance information and supporting software.

Arm says these developments reflect its broader strategy of providing a common computing platform spanning cloud infrastructure, edge devices and physical AI systems such as robots, vehicles and other autonomous machines.

MENAFN23092026005532012229ID1111702439