Alphabet Inc Class CGoogle launches Gemini Robotics ER 2, a new embodied reasoning model that enhances robot perception, planning, and multi-robot cooperation.

Alphabet's Google announced the launch of Gemini Robotics ER 2, a new version of its embodied reasoning model designed to improve how machines perceive, plan, and execute difficult tasks. The model provides a high-level planning framework that allows robots to interpret natural language, understand their environment, and coordinate multi-step activities while lower-level models handle physical movements. It also enhances multi-robot cooperation, demonstrated on Apptronik's Apollo 2 humanoid robot, a Franka F3 Duo robotic arm, and Boston Dynamics' Spot robot. The update includes improved work planning, spatial reasoning, and visual understanding over its predecessor, along with safety features such as stopping a robot when a person enters its workspace and resuming when the area is clear. Developers can access the model via the Gemini API and Google AI Studio, with a private preview available on the Gemini Enterprise Agent Platform.
Alphabet Inc Class CGoogle launches Gemini Robotics ER 2, a new embodied reasoning model that enhances robot perception, planning, and multi-robot cooperation.
Apptronik's Apollo 2 humanoid robot is featured as a demonstration platform, suggesting increased visibility and potential demand.
Boston Dynamics' Spot robot is used as a demonstration platform for the new model, indicating potential integration and demand for its robots.