Teradyne Robotics, the operating arm behind Universal Robots and Mobile Industrial Robots (MiR), said on June 11, 2026 it will stage a wide range of "production-ready" Physical AI applications at Automate 2026, June 22-25 at McCormick Place in Chicago, booth #1250. The lineup is heavy on real deployments rather than concept demos, reflecting an industry-wide push to turn physical AI hype into purchasable applications.
The flagship Physical AI products
The first physical-AI product Teradyne is shipping is MiR's MiR1200 Pallet Jack, available through the group's global system integrator and partner ecosystem. Universal Robots' next-generation PolyScope X software platform anchors the lineup, layering modern web technologies, containerized applications, native ROS 2 support and PLC-style Logic Programs that run in parallel with the main robot program. The booth will host hourly live demos featuring PolyScope X as the tech stack for cyber-secure, AI-ready robotics.
Targeting electronics manufacturing and AI data centers
Teradyne is leaning into the data-center buildout that dominates capex headlines. The UR AI Trainer, co-developed with Scale AI, lets users physically guide a UR robot through tasks like smartphone packaging, capturing high-fidelity force-aware data to train Vision-Language-Action models for direct factory deployment. Operators can pipe that data into NVIDIA's GR00T open VLA model or validate it in NVIDIA Isaac Sim. A second cell uses dual UR12e robots running Generalist's GEN-1 foundation model for dexterous manipulation. A third demo - run with Cambrian - puts dual-arm UR7e robots and a Cambrian vision system on inserting copper cables into high-density server racks, a task notorious for manual errors and labor shortages.
Ecosystem partners take center stage
Partner-led demos round out the booth. AICA's force-sensing buffing cell adapts a UR7e to operator demonstrations after a single example. beRobox's PALTZ palletizer pairs a UR20 with an AI vision redirect system and the MiR600 to handle pallets that have shifted in transit. Maple Advanced Robotics shows operator-marked autonomous spot sanding on a UR8L, and Trener Robotics' Acteris platform lets operators set up new robotic machine-tending jobs via chat input in 1-2 minutes. Vention's Rapid Operator AI bin-picking solution claims a 99% first-pick success rate, configured in minutes with zero programming.
Physical AI hype meets enterprise budgets
Teradyne's pitch dovetails with Nvidia's Unitree humanoid platform announcement and the recent SkildAI $1.4B Series C from SoftBank. Where humanoids capture investor attention, Teradyne argues that mobile cobots delivering boxes and arms inserting cables are the work today's customers can sign POs for.
Reporting based on coverage from Robotics Tomorrow and Teradyne.