首页>
外国专利>
MECHANICAL ARM-ORIENTED NON-KINEMATIC MODEL TRAJECTORY TRACKING METHOD AND MECHANICAL ARM SYSTEM
MECHANICAL ARM-ORIENTED NON-KINEMATIC MODEL TRAJECTORY TRACKING METHOD AND MECHANICAL ARM SYSTEM
展开▼
机译：机械扶手导向非运动模型轨迹跟踪方法和机械臂系统
展开▼
页面导航
摘要
著录项
相似文献
摘要
A mechanical arm-oriented non-kinematic model trajectory tracking method, comprising the following steps: obtain an actual trajectory equation ra(t) of a mechanical arm at a time t by means of a sensor, and combine a predetermined target trajectory equation rd(t) to obtain a first error function e(t); apply a gradient neural network to the error function e(t) to obtain a differential equation (I) for the rate of change of a state of a mechanical arm driver; obtain an actual speed equation (II) of an end effector and a product equation (III) of a Jacobian matrix and a driver rate by means of the sensor to obtain a second error function ε(t); apply the gradient neural network to the error function ε(t) to obtain an equation (IV); jointly solve the equation (I) and the equation (IV) to obtain a joint state vector θ(t) of the mechanical arm; and a controller drives movement of the mechanical arm according to the joint state vector θ(t) of the mechanical arm, thereby completing trajectory tracking. In the case that industrial robot model parameters are unknown, a tracking task is completed only by means of target trajectory information and a measurement result of the sensor, and the accuracy of trajectory tracking can be controlled by changing equation parameters.
展开▼
机译：面向机械臂的非运动模型轨迹跟踪方法，包括以下步骤：通过传感器，获得机械臂的实际轨迹等式R a / sub>（t）的时间t，组合预定的目标轨迹等式R D （t）以获得第一误差函数E（t）;将梯度神经网络应用于误差功能E（t）以获得用于机械臂驱动器状态的变化率的微分方程（i）;通过传感器获得延伸末端执行器的实际速度方程（II）和雅各比矩阵的产品方程（III）和驾驶员速率，以获得第二误差函数ε（t）;将梯度神经网络应用于误差功能ε（t）以获得等式（IV）;共同求解等式（I）和等式（IV）以获得机械臂的关节状态矢量θ（t）;控制器根据机械臂的接合状态矢量θ（t）驱动机械臂的运动，从而完成轨迹跟踪。在工业机器人模型参数未知的情况下，仅通过目标轨迹信息和传感器的测量结果来完成跟踪任务，并且可以通过改变等式参数来控制轨迹跟踪的准确性。
展开▼
