Formulation and Motion Planning of Peg-in-hole Task with Mixed Logical Dynamical System Theory

Abstract— It is well known that most assembly skills can be regarded as one of the signal-symbol Hybrid Dynamical Systems (HDS), since the interactive dynamics between an end effector and the environment change according to contact configurations (geometrical constraints). In this paper, we first propose a model of the peg-in-hole task, which is known as a typical assembly task. MLDS is adopted to formulate the peg-in-hole task, and an optimal solution with both continuous and logical variables under an objective function is found systematically. Then, a strategy to find the optimal solution with less computation is proposed. Finally, some planning results for the peg-in-hole task are shown in order to verify the usefulness of the proposed method.