This work presents advances in the development of an intelligent robotic sanding system for wooden parts manufacturing, focusing on online monitoring of abrasive condition and inspecting surface quality using computer vision techniques. A sanding module equipped with an acoustic emission sensor was implemented to evaluate two abrasive conditions: abrasive loading and wear. The feasibility of using acoustic emission signals to predict sandpaper condition and optimize its replacement strategy was demonstrated. On the other hand, an inspection system based on machine vision algorithms, calibrated through confocal microscopy measurements, was developed to estimate surface roughness parameters. Additionally, the study investigates the relationship between the surface roughness and wettability of robot-sanded parts, which directly affects subsequent finishing processes such as coating, painting, and adhesive bonding. A deeper understanding of this relationship enables to achieve optimal surface characteristics for different manufacturing requirements. The results reported in this work contribute to the development of autonomous sanding cells capable of making decisions based on sensors and data processing techniques in advanced manufacturing environments.