Tool orientation and path optimisation for error compensation in direct robotic plunge milling

Despite the wide success of serial robots in manufacturing applications, precision machining operations are rare, due to their limited motion accuracy. Direct robotic plunge milling, without using external actuation for axial movements, is yet to be investigated and optimised. In a typical plunging process, kinematic and joint reversal errors of the robot offset the cutting tool from its designated trajectory. Additionally, the low stiffness of a serial robot allows the cutting tool to deflect further and can induce self-excited chatter under the cutting force action. Together, these effects may cause significant dimensional errors in both hole positions and bore profiles. This paper identifies these critical error sources during the robotic plunge milling process and proposes a novel optimisation strategy that combines feedforward motion planning and feedback motion control to minimise the errors. The performance of the optimisation strategy is initially evaluated in airborne plunge milling tests for suppressing movement errors of the tool centre point. Aluminium plunge milling tests are then presented to analyse the suppression of dimensional errors in milled holes. The experimental results show that the proposed strategy significantly improves the machining accuracy in direct robotic plunge milling, reducing manufacturing errors by more than 80%.