–

As Tesla’s camera-only approach to autonomy gains traction, Waymo’s EMMA model could mark a major pivot in the industry. EMMA (End-to-End Multimodal Model for Autonomous Driving) abandons radar and LiDAR for a camera-centered, multimodal model built on Google’s Gemini. By leaning into lower-cost visual sensors, EMMA aligns with Tesla’s vision of affordable, scalable autonomous vehicles, potentially pushing the industry closer to a $30,000 price point.

:

• End-to-End Learning: EMMA processes camera data to deliver critical outputs like trajectory planning and 3D object detection, streamlining the driving model.

• Unified Language Space: By using natural language inputs and outputs, EMMA leverages Gemini’s knowledge for greater adaptability and contextual awareness.

• Chain-of-Thought Reasoning: This feature improves planning by 6.7%, allowing EMMA to make reasoned, on-the-fly decisions in complex environments.

• Multitasking Efficiency: EMMA handles multiple tasks trajectory prediction, object detection, road graph analysis together, boosting both flexibility and performance.

By embracing camera-driven intelligence, Waymo’s EMMA is pushing autonomous tech beyond cars, potentially integrating it across devices and wearables. This kind of tech ecosystem could redefine how we connect with our surroundings and each other, creating networks where vehicles, devices, and wearables seamlessly collaborate to support any users needs while maintaining a broader shared context.

“ ’ , ’ , , . , ’ , , . .” –

#MindsetChange #Innovation #ai

: The views within any of my posts, or newsletters are not those of my employer or the employers of any contributing experts. this? Feel free to reshare, repost, and join the conversation.