Kairobotics

The European Union's logistics and warehousing sector faces a significant labor shortage, with 76% of leaders reporting this challenge (Descartes). This shortage is driven by factors including an aging population and declining birth rates (European Commission - Eurostat), insufficient workers with the technical skills required (European Commission), and the industry's perceived unattractiveness (Gigroupholding). The labor shortage contributes to operational inefficiencies, delivery delays, and increased costs. Additionally, the surge in e-commerce and demand for faster shipping are driving a need for higher throughput. The sector also faces limitations with traditional automation in handling complex or variable tasks, which require manual labor. The particular problem we address is automating historically difficult to automate logistics and warehousing tasks.

To address the growing needs within the EU logistics and warehousing sector, we propose implementing (mobile) robotic manipulators powered by Vision-Language-Action (VLA) models. We specifically target the high-variability, labor-intensive tasks that are challenging for traditional automation, such as sorting, packing, mixed SKU palletization, and item picking. These models generalize to diverse situations, requiring only fine-tuning for optimal performance. They offer a flexible alternative to rigid, custom automation solutions. Our approach involves fine-tuning a state-of-the-art foundational model (like π_0) using imitation learning on robotic manipulators. This directly addresses workforce shortages by reducing labor dependence while increasing overall throughput to meet growing demands.

Our mission is to revolutionize logistics by providing intelligent robotic automation for complex warehouse tasks.