Industrial challenge

Automotive interior components combine rigid plastic supports with flexible coatings that must be positioned very precisely on highly configurable lines. Today, operators are responsible for loading parts that arrive in random orientations and for visually checking finished components, including subtle aesthetic imperfections. This combination of complex geometries, frequent product changeovers and high staff turnover increases the risk of misplacement and late defect detection, leading to unnecessary waste and putting constant pressure on human inspectors.

AI-based solution under test

Within the AIRISE framework, PERSICO is piloting a new workstation in its laboratory that couples collaborative robotics with AI vision. A robot equipped with a dedicated multi‑station gripper retrieves both 3D plastic parts and deformable liners from random loading stations, using AI‑based object detection and pose estimation to guide accurate placement on the line. At the end of the process, a vision system with an industrial camera analyses each produced part and classifies potential defects, including difficult‑to‑quantify aesthetic issues, providing a clear decision interface for operators.

The prototype cells are designed for integration into existing and new production equipment and are being evaluated on PERSICO’s internal pre‑series lines. The assessment focuses on cycle time, positioning accuracy, robustness of defect detection and overall cost‑effectiveness, with the objective of enabling deployment to more than 200 customer plants once validation is complete.

Experience within AIRISE

Participation in the AIRISE Open Call has allowed PERSICO and INTELLIMECH to collaborate closely and test the solution under realistic manufacturing conditions. The AIRISE services have supported the refinement of the vision algorithms, streamlined data acquisition and facilitated the identification of critical issues at an early stage. AIRISE project is a valuable opportunity for SMEs and mid‑caps to access targeted AI expertise, benefit from a structured experimentation framework and build a concrete roadmap for scaling AI across their own production lines.