optimization.virtual¶

Virtual states in a plan: a controller’s own states, as unknowns advanced as it advances them.

Functions

`advance`(z, zdot, w)	The virtual state `w` [s] on: the controller's own semi-implicit Euler (velocities first), the same as `VMCController.step` and `vmc.sim.rollout`.
`blended_law`(builder, compiled, hold)	`law(q, v, z, z_old, t, w, p)` gives the motor torques of the controller with weight `w` and of the one in place (compiled, live Params, virtual state) with weight `1 - w`, and the rates of both virtual states, at the state of the plant.
`nodes`(builder, name, start, count)	The virtual state at `count` nodes: unknowns that start from `start`, or empty columns when the controller has no virtual states.
`start_of`(compiled, given, what)	The virtual state a controller starts from: `given` (positions, then velocities), or the state it is compiled with.