Intelligent Robots and Systems (IROS) 2026
I made it! After decades of watching these conferences from afar, never having the funds or time to attend, I can finally say I am here! IROS 2026 is one of the largest robotics conferences for the research community, and thankfully it was located in the US this year.
Below is a list of the presentations, workshops, and tutorials I will attend this week. The breadth and depth of the subject matter are impressive. Though I may not understand it all, it will certainly stimulate new ideas and fill up my reading list.
IROS 2026 — Personalized Conference Schedule
Prioritized by Continuous Learning, AI-SLAM, Kinematically Independent Locomotion, Sim-to-Real, and ROS 2.
| Day & Time | Room | Session / Workshop / Track | Focus & Thematic Alignment |
|---|---|---|---|
| Sun, Sep 27 8:30 AM – 12:30 PM |
Room 315 | 2nd International Workshop on Intelligent Robotics for Soft Tissue Surgery | Continuous Learning & Sim-to-Real: Reinforcement learning, sim-to-real transfer, and online adaptive control in uncertain, dynamic environments. |
| Sun, Sep 27 1:30 PM – 5:30 PM |
Room 412 | The 16th Workshop on Planning, Perception and Navigation for Intelligent Vehicles (PPNIV 2026) | AI-SLAM & Continuous Adaptation: Robust open-world perception, continuous adaptation under dynamic conditions, and learned spatial representations. |
| Mon, Sep 28 8:30 AM – 9:00 AM |
Main Hall C | Welcome Ceremony | Plenary Event: Official conference opening and orientation attended by all registered participants. |
| Mon, Sep 28 9:00 AM – 10:30 AM |
Session Rooms | Contributed Talks: Continuous & Online Robot Learning | Continuous Learning: Policy adaptation during sustained deployment, non-stationary task execution, and mitigating catastrophic forgetting. |
| Mon, Sep 28 11:00 AM – 12:30 PM |
Halls A & B | Poster Session & Paper Demos: AI-Inspired SLAM & Neural Representations | AI-Inspired SLAM: End-to-end differentiable mapping, 3D Gaussian Splatting / neural field SLAM, and dense semantic loop closure. |
| Mon, Sep 28 1:30 PM – 2:30 PM |
Main Hall C | Plenary Panel: Start-ups and Entrepreneurship | ROS 2 & AI Deployment: Commercializing Physical AI, moving from prototype foundation models to production ROS 2 systems. |
| Tue, Sep 29 8:30 AM – 9:00 AM |
Room 405 | Community Building Day (Kickoff & Networking) | Community Session: Early morning networking, community discussions, and student/early-career mentorship. |
| Tue, Sep 29 9:00 AM – 10:30 AM |
Session Rooms | Contributed Talks: Unstructured & Rough-Terrain Locomotion | Kinematically Independent Locomotion: Proprioceptive feedback, blind locomotion over rubble, and morphology-agnostic policies. |
| Tue, Sep 29 2:30 PM – 4:00 PM |
Session Rooms | Contributed Talks: Sim-to-Real Transfer & Verification | Sim-to-Real: Differentiable physics engines, domain randomization bounds, and system identification for zero-shot hardware deployment. |
| Wed, Sep 30 8:30 AM – 9:00 AM |
Session Rooms | Contributed Talks: Middleware & Real-Time AI Systems (Part 1) | ROS 2 in the AI Era: Deterministic executor scheduling and low-latency pipeline designs under heavy model evaluation. |
| Wed, Sep 30 1:30 PM – 2:30 PM |
Main Hall C | Plenary Panel: Generalist vs. Specialist Robotics | Embodiment-Agnostic Models: Generalist cross-embodiment foundation policies versus kinematically specialized robot controllers. |
| Wed, Sep 30 2:30 PM – 4:00 PM |
Session Rooms | Contributed Talks: ROS 2 in the AI Era (Part 2) | ROS 2 & Edge AI: Micro-ROS integration, edge device execution, and zero-copy shared memory pipelines for VLA models. |
| Thu, Oct 1 8:30 AM – 12:30 PM |
Rooms 319 & 320 | 3rd Workshop on AI Meets Autonomy: Vision, Language, and Autonomous Systems | AI-SLAM & Spatial Reasoning: Situated visual-language navigation (VLN), grounding foundation models, and embodied spatial perception. |
| Thu, Oct 1 1:30 PM – 5:30 PM |
Room 415 | 5th Sensorimotor-Augmented Interaction: Human-Interaction-Oriented Paradigm | Continuous Learning & Shared Autonomy: Online policy adaptation, safety filters, and human-in-the-loop sim-to-real calibration. |
Time permitting, I will post short essays on my initial impressions of promising research, particularly if it is available to try as an open-source project.
Stay tuned!
Comments ()