Abstrato

Mathematical model of nonlinear distortion and linear error correction for soccer robot vision system

Chao Chen, Guang Lu, AbdolGhaffar Ebadi

For the soccer robot vision system this paper conducts distortion and error analysis, for the linear part and non-linear part of the systemÂ’ distortion and error it conducts the principle research, uses the geometric similarity theory to establish a linear error correctionmodel, uses theK value selected and the least squares principle to establish the nonlinear distortion correction model, and then uses the linear error correction model and nonlinear distortion correctionmodel in this text to correct the coordinate values for soccer robot vision system. The results show that the kicked rate after a simple linear error correction has increased 18.04% than that before correction, the kicked rate after a simple linear distortion correction has increased 14.32% than that before correction, the kicked rate after composition of linear error correction and linear distortion correction has increased 30.77% than that before correction, showing that the effect of simple linear error correction is stronger than that of simple nonlinear distortion correction, but the superimposed correction effect is the best.