The customer is a leading company in the field of automotive glass manufacturing, which needs to use a robot to pick up the glass, apply glue and install it on the car body. The glass itself is a transparent object with certain reflective characteristics, which makes it difficult to detect, coupled with the large size of the upper window, front glass, etc., the 3D vision system is required to have excellent detection capabilities while taking into account the wide field of view, precision to meet the requirements of the bonding process, and reasonable corrective action for the bonding trajectory so that there can be no distortion, leakage, etc. In this project, a laser 3D camera with a wide field of view is used in combination with algorithm optimisation to meet the customer's glass bonding process requirements.