Is there a name for the bounded piecewise linear function
$$ f(x) = \begin{cases} x,& x\in [-1,1] \\ -1,& x < -1, \\ 1,& x > 1. \end{cases} $$
It resembles the tanh function but does not have continuous derivatives.
On
In the deep learning (artificial neural networks) literature, the function is known as the "hard tanh" function. In their course notes https://cs224d.stanford.edu/lecture_notes/LectureNotes3.pdf for "CS 224D: Deep Learning for NLP", Rohit Mundra and Richard Socher state that "The hard tanh function is sometimes preferred over the tanh function [as an activation function for neural networks] since it is computationally cheaper."
It is a ramped step function.
You may express $$f(x) = \begin{cases} x,& x\in [-1,1] \\ -1,& x < -1, \\ 1,& x > 1. \end{cases}$$
as a heaviside function.
On the positive reals your function is $$f(t)=t-(t-1)H(t-1)$$ and on the negative reals it is $$g(t)=-f(-t).$$