If our DMP system is planning a path for the hand to follow, then what gets sent to the real system is the set of forces that need to be applied to the hand. 首先这里的举例是机械臂的
But keep in mind that the whole DMP framework is for generating a trajectory \ control signal to guide the real system. 所以可以总结DMP的作用就是生成路径或控制信号的
point attractor dynamics 这个到底咋个翻译好？点到点的控制嘛？
$$ \ddot{y}=\alpha_y(\beta_y(g-y)-\dot y) \tag{1} $$
其中
可以看出是一个简单的PD控制器，再次基础上加入forcing term $f$ 是非线性方程，怎样去定义这样一个$f$ 也是一个难题
$$ \ddot{y}=\alpha_y(\beta_y(g-y)-\dot y)+f \tag{2} $$
通过canonical dynamical system $x$的引入，对$f$的定义：
$$ f(x, g)=\frac{\Sigma_{i=1}^{N} \psi_{i} w_{i}}{\Sigma_{i=1}^{N} \psi_{i}} x\left(g-y_{0}\right) \tag{3} $$