From humanoid machines and autonomous navigation to tactile sensing and the future of robotics, Pittsburgh will host one of the field's biggest research gatherings this fall.

Stylized illustration of a glowing robotic figure against a dark background.

The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) is coming to Pittsburgh this fall. It brings researchers, engineers, students, and practitioners together around the systems that allow robots to perceive, move, learn, and work alongside people.

IROS 2026 is scheduled for September 27 through October 1, 2026, at the David L. Lawrence Convention Center.

A major robotics conference, in a robotics city

The 2026 gathering is the 38th edition of the conference.

The official IROS 2026 website says more than 5,000 attendees are expected, with participants from more than 70 countries, and lists 1,500+ paper submissions. IROS 2026 lists Howie Choset of Carnegie Mellon University as its General Chair. Carnegie Mellon University is listed as a Platinum Sponsor of the conference; that sponsorship has no connection to Posh Pittsburgh.

Those details are announced by the event organizers and provide context for the week, but they are not a promise about attendance, access, or the experience of any individual visitor.

Why Pittsburgh makes sense

Carnegie Mellon University’s Robotics Institute was founded in 1979, and the city has been a place where robotics research is tested against practical problems for decades. CMU’s Navlab projects helped make autonomous vehicle research visible to a broad audience, and Pittsburgh’s concentration of robotics companies, researchers, and technical institutions helped earn it the nickname “Roboburgh.”

That history gives IROS a setting with some useful texture. The ideas presented in conference rooms connect to a city that has long used technology to navigate its rivers, hills, hospitals, roads, factories, and changing neighborhoods.

For guests, Pittsburgh can be a living backdrop to the week rather than just a place on a conference badge. A walk along the river, a meal downtown, or an evening in one of the city’s neighborhoods can make the research feel connected to an actual place.

What to expect from the IROS 2026 program

Humanoid robots

Humanoid robots are one of the clearest themes to watch, including the humanoid assembly challenge in which machines must coordinate perception, movement, and manipulation in a task that resembles work in a human environment.

  • Whole-body control
  • Sim-to-real locomotion
  • Loco-manipulation

In plain English, the question is not just whether a robot can move an arm or take a step in a controlled demonstration. It is whether the whole system can balance, see, plan, grasp, recover from small errors, and keep working as the surroundings change.

That last challenge is especially important. A robot that looks impressive in a carefully prepared clip still has to deal with uneven floors, objects in the wrong place, shifting light, and people who do not move according to a script.

Autonomous navigation

Autonomous navigation asks how robots understand where they are and decide where to go. The program’s work in this area can reach from indoor mapping and mobile manipulation to outdoor environments, shared spaces, and systems that must make decisions with incomplete information.

The most interesting demonstrations are often the ones that show judgment rather than speed: slowing down near an obstacle, choosing a safer route, recognizing that a map is out of date, or asking for help when confidence drops.

Machine learning and human-robot interaction

Machine learning connects many of these themes, from perception and prediction to the way robots learn from demonstrations, simulation, and real-world feedback. Human-robot interaction adds another layer: people need to understand what a robot is doing, what it expects, and how to intervene when something goes wrong.

That is where technical performance meets everyday trust. A useful robot will need to interpret human behavior without overclaiming what it knows, communicate its limits, and behave safely in a crowded room.

Competitions can make the research easier to see

IROS 2026 lists nine competitions, including the Volting Cup – Embodied Intelligence, Expressive Humanoids & Inclusive Mobility; The Earth Rover Challenge; The Humanoid IKEA Assembly Challenge; The Robotic Origami Challenge; The 31st RoboRacer Autonomous Racing Competition; the Forceps-based Fine Manipulation Challenge; ROCO: 2nd Robotic Collaborative Assembly Challenge; Legged Robot Challenges; and The 2nd Robotic Construction Challenge.

The exhibit hall and program overview place the exhibition in Halls A and B from September 28–30, Monday through Wednesday, 9 AM to 5 PM. Alongside the exhibit hall, the conference includes keynotes, technical sessions, workshops, plenary panels, and demonstrations.

The announced plenary topics include start-ups and entrepreneurship, space robotics, and the question of generalist versus specialist robots. The program also includes a US government funding forum and a robotics education forum, which broaden the conversation beyond individual machines and papers.

What prior IROS conferences tell us

Recent IROS venues include Prague in 2021, Kyoto in 2022, Detroit in 2023, and Abu Dhabi in 2024, as listed in the conference’s past and future venues. IROS 2022 in Kyoto reported more than 4,000 participants.

For prior-year submission context, the IROS 2025 conference digest reported 4,306 conference paper submissions. That figure describes an earlier conference, not the 2026 event.

That is historical context, not a forecast for 2026. Conference size, submissions, schedules, and participation can change, and the official 2026 site remains the right place to confirm current information.

The audience is likely to include researchers, students, engineers, educators, founders, public-sector representatives, and people who work around robotics without presenting a paper themselves. That range is part of what makes a large research gathering legible to a visitor: the field is technical, but its questions increasingly touch transportation, care, manufacturing, education, and daily life.

What we’ll be watching in 2026

The following are possibilities to watch, not announcements of specific breakthroughs.

Will humanoid robots show useful movement, not just impressive movement?

Prior-year coverage offers a useful point of comparison. Humanoid.Guide’s field notes from IROS 2025 were dated October 22, 2025; Live Science’s coverage of Unitree’s H2 was dated October 21, 2025; and RoboHorizon’s report on the Unitree H2 at IROS 2025 was also dated October 21, 2025. Those reports provide context for the kinds of demonstrations that drew attention at IROS 2025, but they do not establish what will be launched or demonstrated at IROS 2026. The question for 2026 is whether movement that looks striking also becomes repeatable, adaptable, and useful in a real task.

Will touch become as important as sight?

Vision is only one way to understand the world. Tactile sensing can help a robot recognize grip, texture, contact, force, and slippage. At IROS, it will be worth watching how much attention moves from seeing an object to knowing what is happening at the moment a robot touches it.

Will research move closer to everyday environments?

Robots do not work in a vacuum outside the lab. Homes, hospitals, streets, warehouses, classrooms, and public buildings are full of interruptions and people. A key measure of progress will be how research handles those ordinary complications without making the system unsafe or unusable.

Will the field favor generalists or specialists?

There is a continuing tension between a robot designed to do one thing exceptionally well and a more general system designed to handle many tasks. IROS can show where researchers believe adaptability is worth the added complexity, and where a focused machine remains the better tool.

A guest’s view of the week

For a guest, the conference can sit inside a larger Pittsburgh rhythm. A riverfront walk before the day’s sessions can clear the head; Downtown restaurants make an easy dinner plan; and nearby neighborhoods offer a change of pace when the convention center starts to feel like its own small city.

That contrast is part of the story: a week centered on robotics can still be experienced through ordinary Pittsburgh pleasures—walking, eating, noticing the hills, and finding a little room between scheduled sessions.

Official IROS 2026 details

Independent event information: Posh Pittsburgh is sharing information about IROS 2026 as a Pittsburgh guest-planning resource. Posh Pittsburgh is not the event organizer, sponsor, exhibitor, ticket seller, or official partner of IROS, IEEE, RSJ, Carnegie Mellon University, or the David L. Lawrence Convention Center. Event dates, schedules, registration requirements, exhibit information, prices, availability, and policies are controlled by the event organizers and may change. Please confirm current information through the official IROS 2026 website: https://2026.ieee-iros.org/

IROS 2026 is a professional conference, not a free open-public event. Registration or other access requirements may apply.