Kuala lumpur: General Robotics, a company focused on developing an intelligence layer for physical artificial intelligence (AI), has announced significant advancements in its GRID robot intelligence platform. These enhancements aim to automate critical stages of robot deployment, significantly reducing the time and specialized expertise required to transition robots into production environments. According to BERNAMA News Agency, GRID now leverages AI to streamline processes such as robot onboarding, AI model integration, and the creation and deployment of new robotic skills. Ashish Kapoor, the founder and chief executive officer of General Robotics, stated that the platform consolidates robotics knowledge from various domains into a unified system. This integration enables intelligence to build progressively with each deployment. The latest updates to GRID have substantially decreased robot onboarding time from a month to as little as two hours. Similarly, model ingestion time has been reduced from three da ys to 20 minutes. The transfer of skills across different robot types now takes about 1.5 hours, and the creation and deployment of new skills can be achieved in just two days. General Robotics emphasized that GRID employs knowledge graphs to transform each deployment, task, model, and failure into structured, reusable intelligence while safeguarding customer data and intellectual property. The platform is compatible with robots from various manufacturers and form factors, allowing for the selection of systems tailored to specific tasks. Its expanding ecosystem includes partnerships with industrial robotics maker FANUC and bimanual mobile manipulation company Galaxea Dynamics. The company's clientele spans global firms in automotive manufacturing, port operations, energy generation, and food and beverage production, as well as government agencies. General Robotics highlighted that the robotics industry is challenged by a shortage of specialized talent and a fragmented development stack, which necessitates b espoke integration of software, communication protocols, and programming systems, thereby hindering the move from pilot projects to large-scale production deployments. GRID is designed to overcome these limitations by offering a modular library of robotic skills, foundation models, and classical techniques applicable across various robot types and applications. The latest iteration of GRID can identify the necessary skills for a task, select and integrate relevant models, conduct simulations, deploy AI skills, and monitor their performance. The platform then utilizes real-world outcomes and failures to make further refinements, establishing a continuous development cycle across interconnected robots and tasks.