Humanoid Announces its KinetIQ Ascend Reinforcement Learning Approach

London, UK — June 29, 2026 — Humanoid, a UK-based robotics and AI company, has unveiled KinetIQ Ascend, a reinforcement learning approach designed to achieve 99.9% manipulation reliability at human speed and beyond. Building on the company’s existing KinetIQ platform, the new system uses trial-and-error learning to help robots improve directly on industrial tasks, marking a significant step toward deployment-ready robotic capabilities.

Real-World Testing Across Diverse Tasks

KinetIQ Ascend was tested on several industrial tasks, including picking parts from bins, handing objects to humans, and lifting and moving containers using both arms. The system proved effective across a range of manipulation scenarios, demonstrating consistent performance improvements.

In a machine-feeding application where a robot picks steel bearing rings from a bin and places them onto a conveyor, KinetIQ Ascend increased throughput by 42%. The robot operated at 1.5 times the speed of the human demonstrations it originally learned from.

In a task involving picking items from a cluttered tote and handing them to a person, the approach increased throughput by 85% while improving success rates from 80% to 98%. In a more complex bimanual tote-handling task—where the robot lifts a tote using both arms—throughput more than doubled, and success rates rose from 78% to 99%, representing a roughly twentyfold reduction in failures. All results were achieved after only a few days of training.

Scaling Performance Predictably

The company reported that KinetIQ Ascend demonstrates predictable performance improvements as training time increases. This scaling trend, supported by simulation experiments, suggests the method can scale all the way to 100% reliability—similar to how large language models improve with more compute and data.

Two additional findings emerged: improving even the hardest part of a workflow can enhance the entire task, and robots were able to generalize to objects they had not seen during training.

A New "Capability Factory" for Robotics

"The humanoid race is becoming a question of scale, and real-world RL can be a core part of the answer," said Jarad Cannon, Chief Technology Officer at Humanoid. "Robots that once required months of manual tuning are now outperforming human demonstrations within days. KinetIQ Ascend, our real-world RL method, offers a new approach to developing robot capabilities. Instead of spending months collecting data and manually tuning every new skill, we can start with a basic behavior and allow RL to refine it into a deployment-ready capability—a process we describe as building a ‘capability factory,' which marks how humanoid robots move from impressive demos to tools that industry can actually rely on."

The full methodology behind KinetIQ Ascend, including training infrastructure, algorithmic solutions, and deeper analysis of the results, has been published in a new technical report.

About Humanoid

Humanoid is a UK-based robotics company building humanoid robots for industrial use. Founded by Artem Sokolov in 2024, the company brings together over 250 engineers, researchers, and innovators from top global tech companies. All robots run on KinetIQ, Humanoid's proprietary four-layer AI framework designed for real-world deployment. With offices in London, Boston, Vancouver, and San Diego, the company is focused on building commercially viable, scalable, and safe robotic solutions for real-world applications.

The source for this article is https://www.roboticstomorrow.com/news/2026/06/29/humanoid-announces-its-kinetiq-ascend-reinforcement-learning-approach/26785/.