Real-Time System Identification for Impact-Based Part Positioning

Simplified part positioning in manufacturing has been achieved using pushing or tapping actuation in place of more complex pick-and-place systems. However, positioning by impact introduces a new source of uncertainty: interfacial frictional effects of sliding, which can result in an uncontrollable and often poorlypredictable actuation distance. The described work provides a friction-based dynamic model of the sliding distance after impact that is used to predict static friction characteristics. A prototype system is simulated and validation data used to improve the model. A control algorithm is also described, tested and validated over a range of actuated masses