Boston Dynamics and Google DeepMind Integrate Gemini to Enhance Spot's Intelligence

Boston Dynamics' Spot quadruped robot equipped with AIVI-Learning. | Source: Boston Dynamics

Boston Dynamics has announced a partnership with Google Cloud and Google DeepMind to integrate Gemini and Gemini Robotics ER 1.6 into its Orbit AIVI-Learning platform. According to the robotics company, this integration will enable a more sophisticated, intuitive, and powerful AI experience for users.

Industrial environments are highly complex, and the assets managed by Boston Dynamics' customers demand capabilities beyond basic object recognition. The company states that Gemini will introduce enhanced reasoning and adaptability to the AIVI-Learning system.

Boston Dynamics' Spot and Orbit systems can now continuously learn about their deployment facilities with unprecedented depth. This capability, the company explains, enables higher-order reasoning and more complex visual analysis.

DeepMind released the upgraded Gemini Robotics-ER 1.6 model today. This reasoning-first model features improved reasoning and multi-view understanding, specializing in capabilities critical for robotics, such as visual and spatial comprehension, task planning, and success detection. It serves as a high-level reasoning model for robots, capable of executing tasks by natively calling tools like Google Search, vision-language-action (VLA) models, or other third-party user-defined functions.

This year, Boston Dynamics has been strengthening its collaboration with Google DeepMind. In January, the company announced that DeepMind would help enhance the intelligence of its Atlas humanoid robot using Gemini Robotics foundation models.

AIVI now delivers total site intelligence

AIVI-Learning provides insights to multiple stakeholders within an operation, offering a holistic view of site activities.

This enables users to derive value across various areas of their business:

Boston Dynamics states that Orbit AI can adapt as a facility's needs change. Since Gemini comprehends complex concepts from the start, the company can rapidly deploy new, advanced capabilities. Furthermore, AIVI-Learning now includes support for gauges.

AIVI-Learning manages advanced tasks such as 5S compliance audits, measuring sight glass fullness from 0-100%, accurately counting pallets, and detecting pools of standing liquid.

Boston Dynamics sees performance gains and downtime decreases

Boston Dynamics reports that customers can expect improved performance across all visual inspections. By leveraging Gemini's advanced intelligence, AIVI-Learning delivers better baseline performance and higher accuracy for critical tasks, including 5S checks, lever detection, object presence verification, and sight glass readings. Accuracy for key inspections, such as digital display readings, will see significant improvement.

Additionally, the system becomes smarter behind the scenes with Zero-Downtime Upgrades. Boston Dynamics continuously updates and refines these sophisticated AI models in the cloud, which means inspection accuracy improves automatically without requiring end-users to manually update Orbit software or schedule downtime.

Furthermore, AIVI prompts will now feature transparent reasoning, allowing users to see exactly how the model interprets a prompt and the logical steps it takes to reach its conclusion.

More about AIVI-Learning, Boston Dynamics’ core intelligence tool

Boston Dynamics designed AIVI-Learning to be the ultimate visual inspection tool for industrial facilities. The company has made it easier than ever to access this advanced intelligence. Whether an Orbit deployment uses a Site Hub, Virtual Machine (VM), or Cloud hosting, users can connect to the company's servers to utilize these powerful AIVI-Learning models.

By implementing AIVI-Learning, facilities gain access to specialized expert models that continuously evolve. Trained on industrial datasets by leading global researchers, AIVI-Learning delivers genuine site-wide intelligence, according to Boston Dynamics.

Looking ahead, in its next launch, Boston Dynamics will apply AI to Site View. This means Orbit will alert users to issues like puddles, unauthorized personnel, and a growing list of supported assets. Activating AIVI-Learning now will provide immediate access to this feature once it becomes available.

Editor’s Note: Alberto Rodriguez, Director of Robot Behavior for Atlas at Boston Dynamics, will share his insights on the state of humanoid robotics at the Robotics Summit & Expo on May 27-28. The keynote panel, scheduled for 10 a.m. ET on the first day, will also feature perspectives from Pras Velagapudi, CTO of Agility Robotics, and Aaron Prather, Director of Robotics and Autonomous Systems Programs at ASTM International.