Research On The Method Of Optimizing Measurement Poses Based On A Parallel Spindle Head

This dissertation deals with the key issues relevant to geometric error modeling,the robot workspace analysis and measurement configuration optimization method based on the observation indexes for a parallel spindle head which is researched and developed by Tianjin University,combined with the demand of the aerospace industry for high performance parallel CNC processing equipment.The paper aims to put forward a kind of concise and effective measurement poses optimization method,to provide important theoretical and technical support for the robot precision system.The following contributions have been made.With the analysis for the bar length,angles and interference between parts,the reachable space and the flexible working space is calculated.The movement curve is continuous and working performance is excellent in the flexible working space.The geometric error model of A3 parallel spindle head is established based on closed-loop vector chain and first-order perturbation method.This error model takes 24 error sources such as zero error and length error into account,provides theoretical basis for the subsequent error identification and compensation.The sensitivity analysis is carried out in the paper.IOOPS algorithm is proposed to select optimal measurement configurations.The algorithm can optimize a random initial configuration set through continuous exchanging element and eventually achieve the optimal measurement configuration set.Observation indexes(OIs)is introduced into the paper as optimizing criterion.Comparing the result of identification and compensation based on the different measurement noises,the optimal algorithm is proved to be very useful for the parallel spindle head.The outcomes of this paper are useful for the establishment of a system to ensure the geometrical accuracy of the parallel robots in the stages of design,manufacturing and application.