论文部分内容阅读
具有嵌入式视觉的仿生机器鱼的摄像头往往安装在头部,为了获取稳定的图像数据,研究了游动过程中头部的平稳性控制问题.首先,基于牛顿-欧拉方法对仿生机器鱼的水动力学进行建模.然后,基于动力学模型,比较了两种鱼体波模型下的机器鱼头部摆动情况.进一步地采用遗传算法对输入到运动关节的参数进行优化,实现机器鱼头部的最小摆动.最后,在自主设计的具有嵌入式视觉的仿生机器鱼上进行了实验.结果表明,在平稳性控制后,头部的摆动幅度明显减小,采集到的图像的稳定性与连续性有较大改进,但游动速度有所降低.该方法为基于嵌入式视觉的运动控制与任务执行提供了有效保障.
In order to get the stable image data, the head control of the head is studied in this paper.Firstly, based on the Newton-Euler method, the biomimetic robot fish Hydrodynamic modeling.Furthermore, based on the kinetic model, the head swings of the robot fish under two fish body wave models were compared.Furthermore, the genetic algorithm was used to optimize the parameters input to the kinematic joints to realize that the head of the robot fish Finally, the experiment was carried out on a self-designed bionic robot fish with embedded vision.The results showed that after the stability control, the swing amplitude of the head decreased obviously, and the stability and continuity of the collected images But the swimming speed has been reduced.The method provides an effective guarantee for the motion control and task execution based on embedded vision.