Arm launches Total Design for Physical AI and robotics framework
ainews ·
Arm has introduced Arm Total Design for Physical AI and a robotics framework aimed at establishing common standards across automated systems in physical industries like mining, agriculture, manufacturing, and transport. The company highlights a large economic opportunity in these sectors, estimating a $200 billion annual compute market by the 2030s. To reduce engineering fragmentation, Arm is bringing together more than 80 partner organizations from software, hardware, and AI, including Amazon Web Services, Hugging Face, NXP, QNX, Qwen, Siemens, and Unitree Robotics. The initiative targets physical systems that combine AI models, runtime software, compute silicon, sensors, and actuators to sense, reason, and act in real-world environments. Arm also introduced the Robotics Capability Framework, a collaborative attempt to define a shared technical vocabulary for robot capabilities, modeled after SAE levels for driving automation and categorizing robots from reactive to self-improving systems. The framework is intended to standardize system parameters such as latency, compute placement, memory, power, determinism, and safety requirements. Arm Total Design for Physical AI extends an earlier collaborative model used for cloud AI infrastructure, and the company demonstrated it in the automotive sector with partners to allow software development and testing before physical silicon was available. Arm is now seeking broader engineering input to expand the Robotics Capability Framework as physical AI implementations evolve.
ainews