Comau automates picking, handling, and palletizing for Decathlon
THE SO WHAT
Comau’s ROS2-based cobot system with AI vision and modular grippers at Decathlon shows retail logistics is ready for flexible, software-defined automation—not just fixed conveyors. If you run high-SKU warehouses, it’s time to re-evaluate which picking and palletizing tasks can move to cobots without a full facility redesign.
READ THE SOURCE
MORE FROM THE WIRE
Robotics & AutonomySwarmer to acquire Ukrainian UGV maker Ratel Robotics for up to $224M
Software-first swarm control buying a frontline UGV manufacturer for up to $224M is a clear bet that autonomy at scale needs tight hardware–software integration. Defense and industrial buyers should expect more bundled “system-of-systems” offerings—procurement shifts from picking robots to picking autonomy stacks plus fleets.
Robotics & AutonomySpain gives WeRide and Uber its first permit for Level 4 self-driving cars
Spain’s first national Level 4 permit for WeRide, Uber, and AVOMO under the ES-AV framework shows European regulators are ready to move from pilots to structured deployment. If you operate fleets or urban logistics in the EU, start mapping which cities and corridors could open to AV partnerships in your 3–5 year planning window.
Robotics & AutonomySkild AI Taps NVIDIA Physical AI to Teach Robots New Tasks From a Single Video
If Skild AI’s S1 can reliably teach robots new, long-horizon tasks from a single video, the constraint on factory and warehouse automation shifts from integrator time to demonstration quality — that compresses deployment cycles and expands the set of viable use cases. Operators with high-mix, changing environments should start mapping which workflows could move from hard-coded automation to demonstration-based training over the next 12–24 months.
Robotics & AutonomySkild AI hit $100M in run-rate revenue selling robots a brain that learns from one video
Hitting a $100M run-rate in ten months with an S1 model that learns tasks from a single video is a strong signal that generalizable robot brains are commercially viable, not just demos. If you run multi-site operations with repetitive tasks, it’s time to revisit your automation roadmap around software-first control, not bespoke hardware.