/home/cercola/matte_server/DiffusionControl/nonlinear/data_generation/pendulum_lmi.py:65: UserWarning: Solution may be inaccurate. Try another solver, adjusting the solver settings, or solve with verbose=True for more information. prob.solve(solver=cp.SCS, verbose=False) /home/cercola/miniconda3/lib/python3.13/site-packages/numpy/_core/fromnumeric.py:3860: RuntimeWarning: Mean of empty slice. return _methods._mean(a, axis=axis, dtype=dtype, /home/cercola/miniconda3/lib/python3.13/site-packages/numpy/_core/_methods.py:145: RuntimeWarning: invalid value encountered in scalar divide ret = ret.dtype.type(ret / rcount) === CONSTRAINED OC === Constrained optimal control on the pendulum (held-out N=10, actuator |u|≤2.0) PRIOR (quadratic-trained, constraint-unaware): satisfies |u|≤2.0: 0.00 median peak |u| = 12.460 median ISE 0.0366 STEERED toward min ISE=∫θ² s.t. |u|≤2.0 (barrier λ=50.0): satisfies: 0.00 median peak |u| = 7.487 median ISE 0.1078 certified 1.00 ISE REGRET vs per-plant constrained oracle: median +nan% (n=0) ISE-vs-actuator constrained-optimal frontier (steered): u_max mean ISE (satisfying) satisfied 1.50 nan 0.00 2.00 nan 0.00 2.50 nan 0.00 3.00 nan 0.00 INTERPRETATION: a non-quadratic, hard-ACTUATOR-constrained optimal-control problem solved through the differentiable non-linear pendulum: the constraint-unaware prior exceeds the actuator limit; steering enforces |u|≤u_max at minimum tracking error (near the per-plant constrained optimum), certified; a tighter actuator costs more tracking error (the frontier).