A haptic quality control method based on the human somatosensory system

The sense of touch is probably the most complex human sense, because it involves a very large number of sensory receptors spread over the whole body, and takes at the same time full advantage of the human nervous system complexity and power. Although this complexity enables us to perceive the world around us and interact with it, it is also a great source of variability when it comes to controlling the quality of products in a manufacturing environment. Indeed, human subjectivity and preferences, as well as the level of cognitive charge, directly influence the performance of a controller, thus impacting the resulting quality of accepted products, as well as the related economical costs and costumer satisfaction level. We address this issue through three propositions that aim at managing the sense of touch in the context of industrial quality control. (1) The proposed formalization of haptic sensations simplifies the description vocabulary and avoids confusing sensations mix. Nine generic elementary sensations were extracted from a semantic analysis of usual descriptors, and used to define generic haptic anomalies. (2) The modelling of the influence of exploration conditions on haptic perception enables to set common and coherent ways to touch objects. (3) The structuring of the control process is inspired by previous studies in visual quality control. It was applied to the sense of touch and adapted to its specificities. In particular, an intelligent decision method based on the exploration model is proposed, to improve the performance of quality control, and reduce the cognitive charge of controllers for them to focus on perceived sensations. This paper presents the solutions proposed in order to solve the problems associated with the use of the human somatosensory system as a measurement system, while minimizing the variability of quality control decisions.