Advancements in Robotic Inspection: Google DeepMind's Gemini Robotics-ER 1.6

Boston Dynamics’ Spot, a four-legged robot, has been upgraded with the ability to accurately read analog thermometers and pressure gauges, thanks to Google DeepMind’s latest AI model, Gemini Robotics-ER 1.6. Announced on April 14, 2026, this model is designed to enhance robotic capabilities for ‘embodied reasoning’ in physical environments. The collaboration between Google DeepMind and Boston Dynamics aims to improve the efficiency of robotic inspections in industrial settings, such as automotive factories owned by Hyundai Motor Group.

Ars Technica

The Gemini Robotics-ER 1.6 model functions as a high-level reasoning tool for robots, enabling them to plan and execute tasks effectively. This includes complex visual inspections, where robots can interpret various instruments, such as gauges and sight glasses, which provide views into tanks and pipes. The model's advancements are particularly significant in performing inspection duties that require intricate visual reasoning to interpret multiple readings and indicators.

Ars Technica

The introduction of "agentic vision" in the Gemini Robotics-ER 1.6 model marks a notable technological leap. This capability combines visual reasoning with the ability to execute code, allowing robots to create a "visual scratchpad" for inspecting and manipulating images. The performance of the new model has dramatically improved, achieving a 98 percent accuracy rate in reading instruments, compared to just 23 percent with the previous Gemini Robotics-ER 1.5 model. Even without the agentic vision feature, the baseline model maintains an impressive 86 percent accuracy.

Ars Technica

Additionally, the Gemini Robotics-ER 1.6 model enhances "multi-view reasoning," enabling robots to utilize multiple camera streams for a better understanding of their surroundings. This improvement is crucial for tasks that require a comprehensive analysis of complex environments.

Ars Technica

While the advancements in robotic inspection technology are promising, there are concerns regarding the implications of increased automation in the workforce. Critics argue that the deployment of such advanced robotic systems could lead to job displacement in industries reliant on human inspectors. The balance between technological innovation and workforce stability remains a contentious issue.

Ars Technica

Google DeepMind has emphasized the transformative potential of the Gemini Robotics-ER 1.6 model in enhancing industrial operations. A spokesperson noted that the model's capabilities represent a significant step forward in robotic inspection technology, allowing for more efficient and accurate monitoring of industrial environments.

Ars Technica

As the Gemini Robotics-ER 1.6 model is integrated into various industrial applications, further evaluations and trials will likely be conducted to assess its long-term impact on operational efficiency and workforce dynamics. The ongoing collaboration between Google DeepMind and Boston Dynamics will continue to shape the future of robotics in industrial settings.

Ars Technica