Luca Macesanu

[Luca Macesanu @LucaMacesanu]

Don’t miss our poster at #ICRA!
June 3rd 3-4:30 WeI2I.312 and June 5th 10:45-11:45 in the Uncertainty in Open-World Robotics Workshop!

[Luca Macesanu @LucaMacesanu]

Can you guess the object by its sound? 🔊

Humans learn to correlate sight and sound through exploration from infancy. We want robots to learn the same way.

Meet CAVER: the first Curious Audio-Visual Exploring Robot, learning about the world through interactive exploration. 🤖✨

[Luca Macesanu @LucaMacesanu]

For more details, check out our paper and website, Paper: https://t.co/FzAiVeGaa1: https://t.co/w0KR5YIZL7

[Luca Macesanu @LucaMacesanu]

Many thanks to ,
Boueny Folefack, Samik Singh, @ruchira_ray, @babbatem, and @RobobertoMM, without whom none of this would be possible.
Work done RobIn Lab at @texas_robotics @UTAustin

[Luca Macesanu @LucaMacesanu]

CAVER builds multimodal knowledge across environments. In kitchen, garage, & playroom tests, it hit 87% material classification accuracy—beating humans at sound-based action recognition! It can even play tunes by ear on drums & xylophones 🥁🎹.

[Luca Macesanu @LucaMacesanu]

CAVER enables active multimodal data collection via:
1️⃣ 3D-printed tool for consistent audio 🔊
2️⃣ Curiosity-driven policy for high-uncertainty regions🔍
3️⃣ Multi-scale visual-acoustic alignment ⚖️
Making dataset building autonomous and efficient.