Arik Slepyan

MIT ICoN Postdoctoral Fellow
Scalable tactile sensing, sensorimotor learning, and robot manipulation
PhD, Johns Hopkins

[Arik Slepyan @arikslep]

We present OpenTac, an open, full-stack ecosystem for 3D tactile sensing on any robot surface.

Anyone can design and fabricate high-density tactile skins for any part of their robot, then read them out, simulate them, and share data across different sensors.

[Nolan Fey @nolan_fey]

Catch me and @khai_ngx presenting Contact-UAN at #CoRL2026 today! JW Marriott Austin, Griffin (2nd floor), poster #180 from 10:30-12pm

[hosan @hosan300]

Tactile is the missing half of robot data. Just saw OpenTac at CoRL, feels like the timing is right.

[Kushal @kushalk_]

@tylerlum23 and I have been busy bringing Play2Perfect to new robots and simulators! Here’s a sneak peek at two robot setups in IsaacLab and MJLab 👇

Come chat with us at our poster today at 5pm at #CoRL2026!

[Anand Bhattad @anand_bhattad]

We will be live-demoing OpenTac at 1 PM CT in the JW Marriott, second floor, at #CoRL2026

We will also be giving away a limited number of OpenTac samples today after the demo.

https://t.co/H4TWfNCcy1

[Arik Slepyan @arikslep]

OpenTac is live at CoRL today!
https://t.co/JJvAoYgQf4

At the end of the day we will be giving away a limited number of free opentac samples.
Come find us at demo 11 (near Amazon FAR demo)

[Krishna Murthy @_krishna_murthy]

We've been demoing OpenTac https://t.co/knuX1N7RzT at CoRL -- catch it today if you missed us yesterday!

JW Marriott - second floor

We will give away a limited number of sensors to groups working on tactile sensing projects and want to try this out.

Three key features we package in OpenTac:

1. Shear sensing -- 3-axis force estimation

2. High-speed sensing -- for sparse contact, delivers 1000+ Hz readout rates, capturing contacts at millisecond-timescale

3. High Dynamic Range (HDR) sensing -- place a 13 kg and a 13 gm object on the same sensor, and still be able to measure both simultaneously

Join our discord server to remain connected with the team that worked on OpenTac, and for any questions you may have about customizing the sensor to your project/robot!

https://t.co/3sFRJ6hZg9

[Krishna Murthy @_krishna_murthy]

We've been demoing OpenTac https://t.co/knuX1N7RzT at CoRL -- catch it today if you missed us yesterday!

JW Marriott - second floor

We will give away a limited number of sensors to groups working on tactile sensing projects and want to try this out.

Three key features we package in OpenTac:

1. Shear sensing -- 3-axis force estimation

2. High-speed sensing -- for sparse contact, delivers 1000+ Hz readout rates, capturing contacts at millisecond-timescale

3. High Dynamic Range (HDR) sensing -- place a 13 kg and a 13 gm object on the same sensor, and still be able to measure both simultaneously

Join our discord server to remain connected with the team that worked on OpenTac, and for any questions you may have about customizing the sensor to your project/robot!

https://t.co/3sFRJ6hZg9

[Khai Nguyen @ CoRL @khai_ngx]

Ever played FIFA ⚽? Now try 𝘑𝘰𝘨𝘢, where nothing is animated. The humanoid you control runs the exact policies from our real robot, in a physics sim right in your browser.
Dribble, receive, pass, shoot, and top the challenge leaderboard: https://t.co/QRbsEuvZC4

[Changyi Lin @changyi_lin1]

Just started open-sourcing ART-Glove!
https://t.co/zHDerEr8GI

Mechanical files and assembly guide are available now.
PCB, FPCB, code, and video guide are coming next!

I’ll also bring the glove to the CoRL demo session on Monday. Come check it out!

[Mahi Shafiullah 🛫 CoRL 🏠🤖 @notmahi]

I will be at CoRL, presenting Do as I Do, and a yet unreleased demo ✋🤖

My opinion on the future of robotics has shifted drastically over the past year. I believe we're headed towards an inflection point: soon, majority of our robot data will be created without robots (or even humans) in the loop. Why? Because scaling robotics at the speed of robots or humans is almost always going to be slower than scaling at the speed of compute.

We want robots that work in the real world. How do we stop making the real world our bottleneck? Looking forward to sharing my thoughts, and would love to hear yours if you are going to be in Austin next week.

[Homanga Bharadhwaj @ CoRL @mangahomanga]

Thanks to @arikslep we’ve added High dynamic Range (HDR) tactile sensing to OpenTac.

The same sensor can now pick up light touches and heavy loads at the same time.

Just update the firmware: no need to change the hardware!

[Krishna Murthy @_krishna_murthy]

Great to see more work along the lines of what we built in OpenTac-CAD https://t.co/Af3zV1042U

The tactile sensing community needs more of this!

@binghao_huang would be nice to figure what a good manifest standard would be, for interoperability across sensors (e.g., OpenTac, FlexiTac, and more)

[Arik Slepyan @arikslep]

Note: the gram readings in the video use a coarse calibration. See post 4 for measurements against a reference scale.

[Arik Slepyan @arikslep]

One of the biggest problems people talk about with tactile sensors is their dynamic range. Most sensors either saturate quickly or aren't sensitive enough to detect light touches.

Why can't we have both?

Introducing HDR tactile sensing with OpenTac.

[Arik Slepyan @arikslep]

HDR is available now as a firmware update for the OpenTac 60-pin readout board and, like the rest of OpenTac, it is open source.

Documentation: https://t.co/QkF0M7c2yE
Code: https://t.co/9duntqVIY3

[Homanga Bharadhwaj @ CoRL @mangahomanga]

What are the right failures to collect for training robot reward models?

We analyze failures from kinesthetic teaching, policy rollouts, human videos, and real2sim and their impact on reward quality and downstream real-world RL.

https://t.co/ijYtJuRDfZ 🧵

[Arik Slepyan @arikslep]

@WickeDStereO_XD @geetachandra21 Dynamic per pixel gain adjustment :)

[Jiawei Fu @JiaweiAcademic]

Impressive dynamic range for tactile sensing! Resolving forces differing by over 1,000× 🚀 within the same frame, without any hardware modifications.

[Geeta Chandra Raju @geetachandra21]

Our OpenTac doesn't take forces lightly, Or heavily, It takes both at the same time!

https://t.co/RgDdzbkwiK

[Younghyo Park ✈️CoRL @younghyo_park]

Robot arms can be extremely heavy compared to the objects they are asked to manipulate, especially with their metal links and highly geared motors.

But is that actually a problem for robot learning, or are learning algorithms powerful enough to handle it? Are we paying any hidden costs for using heavy robots? Here's what we found 🧵

[Anand Bhattad @anand_bhattad]

Update: we can now do high dynamic range (HDR) tactile sensing with OpenTac: https://t.co/EECPbaepgn

[Krishna Murthy @_krishna_murthy]

If you're at @corl_conf next week, come see this (and much more) in action on Monday and Tuesday during the demo sessions

[Krishna Murthy @_krishna_murthy]

Since releasing OpenTac, we've started getting back to a long list of "here's cool features we can add on top".

First up, we have High dynamic range (HDR) tactile sensing! On the same sensor, you can add a 12 g and 12 kg weight side by side, and read them both!

We (mostly @arikslep) designed the readout boards and hardware flexibly enough that we can support this w/o any hardware modifications

Check out https://t.co/bJyCHQ8eE3