The irregular surface morphology of the turtle shell can be applied to the design of the floating plate of a rice transplanter to mitigate the serious problems of mud and water resistance. An adult Brazilian turtle was selected as the research object in this study. A 7‐axis absolute arm measuring machine and Geomagic Studio software were used to acquire and process point cloud data of the turtle’s plastron. Based on the structural characteristics of the turtle shell, four curved surfaces with dense point cloud distributions were segmented, filtered using CATIA software, and exported as three‐dimensional coordinate data. MATLAB was used to perform polynomial fitting of the three‐dimensional point cloud data. The fitting equations for the four surfaces, as well as the sum of squared errors (SSEs), root mean square error (RMSE), and coefficient of determination ( R 2 ), were obtained. The results showed that the maximum relative errors between the fitted and actual values for the front, rear, and side models were 8.96%, 9.36%, and 5.86%, respectively. The corresponding mean relative errors were 4.58%, 4.67%, and 2.98%, respectively. These mean relative errors fall within the ±5% tolerance permitted in engineering design, thereby verifying the validity of the models and enabling the transformation of the turtle plastron surface from a biological form into a mathematical model. This study provides a theoretical foundation for the bionic application of the turtle plastron surface morphology and offers a reference for the bionic design of floating plates for rice transplanters.
Fu et al. (Thu,) studied this question.