Pedro Lopes
@pedrolopes
Associate Professor at University of Chicago, Computer Science. Runs the Human-Computer Integration Lab () & also a musician.
I published the keynote I have at the Taiwan Human Computer Interaction conference: * Video: www.youtube.com/watch?v=G5Ij... * Drawings from attendees at TAI CHI: vision.plopes.org/taichi * Slides from my talk: lab.plopes.org/taichi
Two-Millisecond Passthrough Headset for Perceptual Studies By Meta's Eric Penner, Josephine D'Angelo, Clinton Smith, Nathan Matsuda, Navaneethan Siva, Phillip Guan "we find that 2 and 14 ms latencies are preferred over 23 and 29 ms when attempting to catch a ball"
I am writing a small reflection about serving as technical program chair for CHI. If you have questions that I can try to answer in that essay (I already got a bunch in instragram) send them my way (you can DM, you can email me annon, reply below—whatever and I'll try to incorporate it!).
What does it look like: Linear resonant actuator (LRA, same vibration as in your phone) provide a single dimension of asymmetric vibration = feels like a force in a direction. How do we get multi directions for 2D cues? We rotate the LRA to where the robot feels forces! 2/3
New paper at #ICRA2026 with/led by Matt Walter's team at TTIC. HapCompass: wearable haptic device for teleoperation that gives directional cues via asymmetric vibration! By mapping the robot's tactile measurements—improves teleoperation performance + better imitation learning!1/3
Letting an AI control my muscles (for science, and yes, it's a research prototype not a product!). Work done by Yun Ho and Romain Nith and published at ACM CHI (best paper award). All details at embodied-ai.tech
I had way too much fun making an example for my class: this is an FPGA driving a screen directly over VGA, no drivers, no adapter, nothing else: just pure H sync and V sync fun ;) almost a game of pong!
Reflection from latest conference our lab attended (ACM CHI): lab.plopes.org/essays/chi20... which includes all the talks we gave, demos, seeing our alumni, catching up with friends, fixing broken demos, and other fun moments we had at the conference!
The most interesting question some folks asked me at #CHI2026 was: "did you not do a video with all the lab's paper this year?" Indeed, given the chairing work, I wasn't able too, but the enthusiasm of these folks helped me do it now 🙏! Here you go: www.youtube.com/watch?v=4VVt...
We did it! Students final boards from hcipcb.plopes.org class are on their way to @JLCPCB . As soon as they arrive it's solder and debug time for them: good luck and enjoy your 1st electronics adventure with a cortex MCU! Super start TAs: Dylan, Yun Ho and Romain Nith
A year ago we made seeingwiththehands.com lead by @tengshanyuan.info , Gene S-H Kim and Xuanyou Liu! On a personal note, it is insane to see how much has changed in just 365 days: Shan-Yuan is a professor now, Gene is a PhD at MIT, Zed is a PhD at Northwestern (and I survived CHI organizing).
Want to ask question to our keynote speakers, use our slido: app.sli.do/event/eYDGwP... #Chi2026 please RT
Want to ask question to our keynote speakers, use our slido: app.sli.do/event/eYDGwP... #Chi2026 please RT
So many conferences in the @acm-sigchi.bsky.social family. Amazing to see this in one slide!!! (Town hall just started at #CHI2026 with kick off by @nehakumar.bsky.social come to the room!)
If you missed the talk (room full, I know!) then you can watch it here: www.youtube.com/watch?v=vd-E... And read the paper here: lab.plopes.org/published/20...
@tengshanyuan.info (now Prof. NTU) new paper is an **argument** we wrote together on why #Haptics Should Balance Virtual & Real-world Fidelity — see it at P1-133 "Embodied Interaction" session. Paper: lab.plopes.org/published/20... p.s. His last contribution while working my group until last year
Today at #CHI2026: 9AM Next Generation Wearable Haptics Should Balance Virtual & Real-world Fidelity—argument paper w/ @tengshanyuan.info at P1-133 11AM MyoAction: Accelerating Voluntary Actions via Electromyography and Muscle Stimulation—new work by Yudai Tanaka at M2-M211
Must watch talk from @jonfroehlich.bsky.social on stage to receive his societal impact award from @acm-sigchi.bsky.social
So what if we correct for the overshoot? Would it lead to better force recalls? Could it lead to better EMS systems that teach you physical skills (e.g., embodied AI systems, tutorial systems, etc). Yes, it can.
How hard is it to learn a force you felt by muscle stimulation? And could we ever use this to build systems that teach you new physical skills requiring controlled forces? Our master student Mithil Guru is presenting this at #CHI2026. Tue 9AM "Haptics + XR" session at P1-119 at @chi.acm.org
That time of the year where Romain Nith and I teach PCB (w @kicad.org and @jlcpcb.bsky.social ) to folks with no electronics/EE background, it's awesome it is to see them make physical devices from scratch! This is hcipcb.plopes.org class at @uchicagopsd.bsky.social @uchicagocs.bsky.social
I'm still so mind blown by all her work I shared a screenshot of Instagram rather than the actual photo... ! Congratulations @xjasminelu.bsky.social
I can't put this into any other words except that "you should check our Dr. Jasmine Lu's work at jasminelu.site " .... A huge congratulations to @xjasminelu.bsky.social on her absolutely stellar PhD and to Eric Paulos Ken Nakagaki and Greg Abowd for serving on this most incredible committee!
But how do people perceive having their bodies moved by an AI? In a study, we found participants successfully completed physical tasks while guided by generative EMS, even when EMS instructions were (purposely) erroneous! (we injected some cool errors) 6/7
Here, the same "help me" request generates two different movements because, a spray can with paint should be shaken, but a spray can with oil needs no shake: the multimodal AI gets this entirely from context (i.e., sensors, POV image from smart glasses, etc) = no custom code! 5/7
To implement our form of Embodied AI we use computer-vision + large-language-models + lots of user-worn sensors (smart glasses, GPS, etc) to generate contextually-relevant EMS instructions, constraining these to a muscle-stimulation knowledge base that cares of joint limits! 4/7