Raunaq Bhirangi

Making an app was uncharted territory for us as a research group, and would not have been possible without my wonderful collaborators: Zeyu (Michael) Bian, @venkyp.bsky.social, @haritheja.bsky.social, @eneserciyes.bsky.social, @notmahi.bsky.social and @lerrelpinto.com.

GitHub - NYU-robot-learning/AnySense: An iPhone app for multi-sensory data collection and learning An iPhone app for multi-sensory data collection and learning - NYU-robot-learning/AnySense

AnySense is built to empower researchers, engineers, and developers with better tools for sensor-based AI. Our code is fully open-source and can be found on

Github: github.com/NYU-robot-le...
Website: anysense.app

Why does this matter? Here, we use AnySense to scale data and train visuo-tactile policies using Robot Utility Models (robotutilitymodels.com) for a whiteboard erasing task. With AnySense-enabled live streaming, you can just plug your iPhone into your robot and seamlessly deploy your policies!

Need to connect external sensors? No problem! AnySense supports data streaming over Bluetooth at the press of a button! Here’s a visualization of data collected by connecting the AnySkin (any-skin.github.io) tactile sensor with AnySense via Bluetooth.

‎AnySense ‎AnySense is an open-source iPhone app that enables multi-sensory data collection by integrating the iPhone’s sensory suite with external sensors via Bluetooth and wired interfaces, enabling both offl...

Try AnySense right now: apps.apple.com/us/app/anyse...

Even if you don’t have external sensors, you can start using AnySense immediately to record and stream:
✅ RGB + Depth + Pose data
✅ Audio from the iPhone mic or custom contact microphones
✅ Seamless Bluetooth integration for external sensors

Ever struggled with multi-sensor data from cameras, depth sensors, and other custom sensors? Meet AnySense—an iPhone app for effortless data acquisition and streaming. Working with multimodal sensor data will never be a chore again!

P3-PO is a great example of how simple human priors can facilitate significantly better generalizability for robot policies.

BAKU is fully open source and surprisingly effective. We found it easily adaptable for a host of visuotactile tasks in visuoskin.github.io

Robot utility models are not just among the first learned models that work zero-shot on a mobile manipulator, but also provide a nuanced discussion on what works and what doesn't in data-driven robot learning.

I'll be presenting AnySkin at the Stanford Center for Design Research today at 2pm! Stop by for a chat and try the sensor out!

More info: any-skin.github.io

I just joined bluesky, and would love to connect with folks interested in embodied AI and robotics. I am a postdoctoral researcher at NYU working at the intersection of sensing, machine learning and robotics. Hit me up if you'd like to chat!

More about my research: raunaqbhirangi.github.io