| The tele-operated construction robot is construction machinery, which is controlled by people, working in extreme environments. In order to ensure that the tele-operation system has good control characteristic of force feedback, the accurate and real time force feedback should be achieved. Therefore, this thesis develops a tele-operated construction robot system, and experimental researches on step response characteristic, position follow characteristic and force feedback characteristic of the system are performed. A force feedback control method based on regression model was proposed to improve force feedback control characteristic of the system. And the usefulness of proposed method is confirmed by experiment. The main research of the paper is as follow.1. The holistic design scheme of the tele-operated construction robot system is determined. The fork glove structure of the construction robot is designed by analogy. A excavator is modified to be the construction robot based on similarity principle. The parameters of the hydraulic system of the construction robot are designed. And the electronic control system of the tele-operated construction robot is designed too.2. The corresponding control software system is developed in the VC++Environment by means of MFC based on Windows operation system. The communications between the joysticks and the computer are developed using DirectX programming interfaces.3. A PID series compensator controller on the slave of the system is devised. Experimental researches on step response characteristic and position follow characteristic are done to improve the position control effect of the system.4. The data of kinetic parameters and no-load driving force of the fork glove are obtained by running the construction robot on no-load condition. According to these data, the influence of the kinetic parameters acting on no-load driving force is analyzed on the Sigma Plot platform and the multivariate nonlinear functional relationship between them is assumed. And then, Taylor’s expansion of multi-function is used to approximate it. Moreover, the optimal regression equations and data about them are obtained by nonlinear regression fitting and the multivariate nonlinear regression model of no-load driving force of the fork glove is built up finally.5. An improved indirect force feedback control method based on the regression model is designed. Experimental researches on force feedback control characteristic are carried on using the novel method. Comparing with the experiment of the conventional indirect force feedback control method, we learn that the improved control method increases the accuracy and synchronization of force-feedback, improves force feedback control characteristic of the system. And then, meanwhile, correctness of the assumed functional relationship and the regression model is verified.The achievements of this study provide the beneficial reference for improving the force feedback control characteristic of the tele-operated construction robot. |
PDF Full Text Request