General Robotics has expanded its GRID robot intelligence platform to automate more of the engineering process involved in onboarding robots, integrating artificial intelligence models, creating skills and deploying them in production environments.
The company said GRID now manages processes across the robot development lifecycle, including simulation, training, testing, deployment and real-world evaluation. The platform is designed to reuse information generated from previous deployments to support subsequent robot and task configurations.
General Robotics said GRID uses knowledge graphs to organize information from robots, tasks, AI models and operational failures. According to the company, the resulting data can be reused across deployments while customer data and intellectual property remain protected.
The company reported that robot onboarding can now be completed in as little as two hours, compared with about one month previously. Model ingestion can take as little as 20 minutes, down from three days, while transferring a skill between different robot form factors can be completed in as little as 1.5 hours, compared with three days. General Robotics said new skills can be created and deployed in as little as two days.
GRID combines reusable robotic skills, foundation models and conventional robotics techniques that can be applied across connected systems. For a given task, the platform determines which capabilities are required, selects a combination of models and methods, and builds simulation environments used to train, evaluate and refine the resulting skill.
The system also manages deployment and monitors robot performance and failures after a skill is introduced. Information from those deployments is fed back into the platform to guide further testing and changes.
General Robotics said its customers include companies in automotive manufacturing, port operations, energy generation and food and beverage production, as well as government agencies. The company did not identify those customers.
GRID supports robots from multiple manufacturers and form factors. General Robotics works with industrial robotics supplier FANUC and robot developer Galaxea Dynamics, among others.
“As a FANUC Authorized System Integrator (ASI), General Robotics is advancing frontier physical AI across the FANUC portfolio — unlocking new AI-driven use cases and greater value for customers,” said Ryan Patterson Jr., general manager for sales, engineering and operations at FANUC America Corporation’s Robotics Division.
Gao Jiyang, founder and chief executive of Galaxea Dynamics, said the company is working with GRID on robotics deployments for customers in industries including automotive, pharmaceuticals, consumer packaged goods and apparel, as well as government organizations.
General Robotics founder and chief executive Ashish Kapoor said the company developed GRID around an agent-based architecture intended to combine robotics research, software, AI models, simulation, data and hardware within a single platform.
The company said its approach is intended to address the engineering complexity involved in integrating different robot hardware, software tools, communications protocols and AI systems. GRID is designed to allow skills and models developed for one deployment to be adapted for other robots and applications. General Robotics is backed by investors including Construct Capital, Accenture Ventures, E14, NVIDIA, Shorooq, Valo Ventures and Khosla Ventures.
