Robotic Learning for Increased Productivity: Autonomously Improving Speed of Robotic Visual Quality Inspection

Abstract

Robotic learning has shown many impressive achievements in laboratory environments, but it is not yet very prominent in the industry, where hardware, engineering solutions and careful structuring of the environment are typically needed to successfully accomplish the desired task. In this paper we show one example of learning applied to industrially relevant problems, where a learning algorithm is applied for the optimization of the velocity of robotic motion for quality inspection. Through learning of optimal velocity of motion, which is done only at the start of the production, we show that we can achieve faster cycle times and thus greater productivity. The described approach is general and can be used with different types of learning and feedback signals.