英文摘要 | With the support of National Hi-Tech Research Development Program and National Natural Science Foundation of China, the research on mechanisms and intelligent control of a biomimetic robot fish has been carried out. In this thesis, the methods are presented, which focus on design of a 3D swimming robotic fish, dynamic models of different motions such as swimming forward, pitching, turning, rising and submerging, parameter optimization of caudal mechanism and intelligent controller for posture control and depth maintaining. Several experiments are done in a small pool, and the results are given. Firstly, The main research directions and progress of both fish’s propulsive mechanism, robot fish’s development and control are reviewed. Secondly, a biomimetic robot fish with the ability of 3D locomotion and suspension is given. According to this design, several mechanisms are realized: a two DOF pectoral fin mechanism which can make pectoral fin tips achieve 8-shaped locus; a caudal mechanism based on fish body wave fitting with one driven motor; a two motor driven caudal mechanism based on the oscillating fin theory; a piston mechanism which has a similar function as gas bladder; and a shell designed under the consideration of the robot-fish’s stability. Thirdly, the dynamics of the robot fish is analyzed. On the basis of the anlysis results, several dynamic models for different motions such as swimming forward, pitching, turning, rising and submerging are established. Fourthly, an optimal method for the caudal mechanism with two driven motors is presented. The optimization process is divided into two parts: the kinematic parameter optimization, which can bring the maximum thrust force based on caudal fin oscillating rule; the mechanical parameter optimization for caudal mechanism, which is based on Genetic Algorithm with the above kinematic parameters and the given motor power. Fifthly, the control methods are presented for robot fish motion control. Because the CPG model can produce a series of rhythm signals similar to the rhythm oscillating of fish’s pectoral fins, two CPG based posture controllers are presented for pitching and yawing control. And a depth controller with fuzzy rules and reasoning is designed, which can make the robot fish go fast and suspend at a given depth by means of adjusting pectoral fins’ attack angles and robot fish’s volume. Sixthly, some experiments such as swimming forward, turning, rising and submerging are done to show the validity of the robot fish prototype. Finally, the research results in the thesis are summarized and future work is addressed. |
修改评论