Robotics Experts Are Watching the Same Big Shift in 2026

May 28, 2026

The robotics industry enters 2026 with momentum, investment and public attention — but also a practical test. As researchers and business leaders prepare to gather for the IEEE International Conference on Robotics and Automation (ICRA) 2026 in Vienna this June, the field is shifting from proving what robots can do in demos to showing where they can deliver reliable value in the real world.

To understand where that test is taking the industry, we asked three robotics experts what trends they are watching most closely this year. Their answers point to a field that is moving beyond eye-catching demonstrations and toward harder questions: how robots learn in simulation, adapt to messy human environments, coordinate movement and manipulation and process intelligence onboard.

Building AI Into Machines

IEEE Senior Member Shaoshan Liu is focused on embodied AI — a term for artificial intelligence that is built into physical machines, allowing them to sense, move and act in the real world rather than only generate text, images or predictions on a screen. In robotics, that shift matters because intelligence has to be matched with a body: the robot’s shape, sensors, computing power, energy use and physical environment all affect what it can actually do.

“I see the field moving beyond the idea that all robots should converge toward one dominant form factor, such as humanoids. Instead, embodied AI will likely evolve through scenario-driven diversification, where different robot embodiments emerge for different tasks, environments, cost structures and data pipelines.”

The Transition to Real World

IEEE Graduate Student Member Dipam Patel is watching the gap between what robots can do in controlled settings and what they need to do in real homes, workplaces and public spaces. His focus is on robots that can understand unstructured environments, move through them and physically interact with them. That includes home robots that can respond to natural-language instructions

“Automated and preprogrammed industrial robots have been in use for several decades now. A regular household is highly unstructured, dynamic and populated by humans, making pre-programmed routines useless for home robotics. To be genuinely helpful with chores like folding laundry, loading a dishwasher or tidying a room, robots must possess a semantic understanding of their environment. “

Managing Drone Traffic

IEEE Senior Member Ayesha Iqbal is watching the development of unmanned aerial vehicles as drones move closer to routine commercial use. Her focus is on improving the communications systems, particularly antennas, that drones need to operate in complex environments. As drones become more autonomous in fields like agriculture and delivery, they’ll need stronger onboard intelligence and more reliable connectivity.

“Drone development in 2026 is focused on making drones more autonomous, efficient and commercially useful, with advances in AI-powered navigation allowing drones to fly complex routes without human control,” Iqbal said.

How Robots Understand the World

IEEE Member Ning Hu is focused on the deeper infrastructure robots will need as they become more autonomous. That includes better 3D representations of physical spaces, faster reflex-like motor skills and tighter coordination between vision and touch. Hu is watching how robots move from following instructions to understanding spaces, reacting quickly and working safely alongside other machines and people.

“Most current robots are still at the stage of deliberate logical inference and lack the kind of reflexive actions seen in biological organisms that do not require high-level cognition. The goal is to enable true embodied intelligence by decoupling high-level task planning from low-level motor primitives.”

Learn More: For anyone trying to understand where robotics is headed next, ICRA 2026 offers a rare look at the research, engineering and commercialization challenges that will shape how robots move from lab breakthroughs to real-world impact.