Google DeepMind has launched Gemini Robotics-ER 1.6, an advanced AI model designed to give industrial robots enhanced capabilities in embodied reasoning, spatial understanding, and task planning. This upgrade significantly improves the robots' ability to interpret complex data, such as reading gauges and sight glasses, a crucial development for industrial automation.
In safety hazard identification tests, the new model shows a 6% improvement in text-based scenarios and a 10% improvement in video-based scenarios compared to its predecessor, Gemini 3.0 Flash. These enhanced features are now accessible to developers via the Gemini API and Google AI Studio.
Boston Dynamics has integrated Gemini Robotics-ER 1.6 into its Orbit AIVI-Learning platform. This integration went live for customers on April 8, marking a direct pathway for practical, real-world application of the AI advancements. Marco da Silva, VP and GM of Spot at Boston Dynamics, stated that these capabilities will allow robots to operate more autonomously in challenging environments.