TransformationsHub management system - Logo

To display the translation please enable cookies for

Awarded the “Technological Innovation” award at the Robotics Challenge 2026.

Mowito: Learning-based robotics for adaptive assembly processes

Mowito's entry for the Robotics Challenge 2026 introduces a physical AI approach to wiring harness assembly. The focus is on the ability of robotic systems to learn complex assembly tasks from demonstrations and adaptively execute them under real production conditions.

Result and solution approach

Mowito's contribution applies learning-based robotics to industrial assembly tasks in wiring harness manufacturing. The goal is to fundamentally simplify the programming of complex robot processes. Instead of programming motion and handling tasks entirely manually, the robot system learns them through demonstrations and data-driven learning methods.

The solution combines methods from imitation learning, reinforcement learning, computer vision, and real-time robot control. This enables the robot not only to reproduce assembly tasks but also to perform them robustly and adaptively under changing conditions. This significantly reduces engineering effort and shortens implementation times and operating costs.

Why this approach is relevant in practice

For wiring harness manufacturing, this approach offers great potential, especially with high product variety and complex handling tasks. The demonstrator shows how adaptive robot systems can make future automation solutions more flexible while significantly reducing the effort required for programming and commissioning.

About Mowito

Mowito develops AI-based software for industrial robotics, focusing on learning and adaptive automation solutions. Founded in 2024 and headquartered in Detroit, the company combines artificial intelligence methods with practical robotics for industrial manufacturing environments and is already deployed in several production lines of leading manufacturers.

For more information about mowito, please visit: www.mowito.ai

Results video of mowito's solution approach
The winning team from mowito at the Robotics Challenge 2026
Contact Person

Robert Süß-Wolf
Research Coordination Management Set

robert.suesswolf@arena2036.de
Contact person

Alexandra Popa
Knowledge transfer and science communication