Embodied AI Breakthrough: RL Boosts Industrial Robot Throughput by 85%
lukas_m_ziegler · x · 2026-07-30
TheHumanoidAI recently presented KinetIQ Ascend, an approach enabling robots to improve themselves via reinforcement learning (trial and error) during real-world deployment, moving beyond mere human demonstrations.
Data indicates that robot performance scales predictably with increased training time and compute, mirroring the scaling laws of large language models. The results in industrial tasks are striking:
- Machine feeding: Picking steel bearing rings and placing them on conveyors saw a 42% throughput increase, operating at 1.5x the speed of human demonstrations.
- Cluttered picking: Grabbing items from a tote to hand to a person jumped 85% in throughput, with success rates improving from 80% to 98%.
More from Embodied
- Autonomous vehicle demo shows car reversing to break narrow-road deadlock, with a robotic arm to shift gears — sanjeevs_iitr · 2026-10-16
- Surface Laptop Ultra review: a beautiful $2,599 laptop for an AI future that hasn't arrived — tomwarren · 2026-10-16
- Waymo brings robotaxis to Toronto, targeting rider service in 2028 — CSProfKGD · 2026-10-16
- Musk shows Cybertruck hovering on jets, delivering on 'flying cars' promise — XFreeze · 2026-10-16
- Unitree Turns Spring Festival Gala Performance into UniUni Robot Swarm Choreography Product — CyberRobooo · 2026-10-16
- Tesla FSD (Supervised) approved in Slovakia, the 9th EU country — XFreeze · 2026-10-16