Abstract:Based on inchworm locomotion principle,a 3-DOF miniature endoscopic inspection squirming robot system is designed,which is driven by a pneumatic rubber actuator and clamped by two air chambers.Dynamic model of the robot is built.With BP neural network PID scheme,an electro-pneumatic PWM(Pulse-Width Modulation)servo controller is designed to control the robot movement.With the predicted value of system output instead of measured value,the weight co-effient is corrected and the controller parameters are changed in real time to improve the control performance.Both simulation and experimental results indicate that this scheme can overcome the shortages of traditional PID controllers,improve the system static and dynamic performance,and is an ideal control method for pneumatic miniature squirming robots.
[1] Adam J A.Medical electronics[J].IEEE Spectrum,1994,31(1):70-73.
[2] Adam J A.Medical electronics[J].IEEE Spectrum,1995,32(1):80-83.
[3] Astrom K J,Hang C C,Persson P,et al.Towards intelligent PID control[J].Automatica,1992,28(1):1-9.
[4] Suzumori K,Iikura K,Tanaka H.Development of flexible microactuator and its applications to robotic mechanisms[A].Proceedings of the IEEE International Conference on Robotics and Automation[C].Piscataway,NJ,USA:IEEE,1991.1622-1627.
[5] 米智楠,钱晋武,龚振邦,等.内窥检查系统主动引导驱动技术的研究[J].高技术通讯,2001,11(6):70-71.
[6] 刘鸿文.材料力学[M].北京:高等教育出版社,2003.
[7] 吴家龙.弹性力学[M].上海:同济大学出版社,2001.
[8] 徐炳辉.气动技术基础知识[J].液压与气动,1994,1:39-41.
[9] Tian S P,Ding G Q,Yan D T,et al.Nonlinear controlling of artificial muscle system with neural networks[A].Proceedings of the IEEE International Conference on Robotics and Biomimetics[C].Piscataway,NJ,USA:IEEE,2004.56-59.
[10] Omatu S,Iwasa T,Yoshioka M.Skill-based PID control by using neural networks[A].Proceedings of the IEEE International Conference on Systems,Man and Cybernetics[C].Piscataway,NJ,USA:IEEE,1998.1972-1977.
[11] Iwasa K,Morizumi N,Omatu S.Pressure control in a plant generating chloride by neural network PID control[A].Proceedings of the IEEE International Conference on Neural Networks[C].Piscataway,NJ,USA:IEEE,1995.627-630.