inline double algorithm_call(std::span<double const> xs) noexcept {
return std::accumulate(
xs, 0.0,
[](double acc, double volts) {
auto mv = calibrated_mv(volts);
auto err = residual(mv);
return weighted_square(err) + acc;
});
}
It's not as easy as it sounds to put the weight of ordnance required in the very tight windows that would be needed to actually cause more than cosmetic or minor damage to that bridge.
Ukraine did pull off a spectacularly successful operation to destroy a laden fuel truck while it was crossing the bridge (https://en.wikipedia.org/wiki/2022_Crimean_Bridge_explosion)... Russia repaired the damage within months and simply redirected fuel transport to other means.
Russia also has dedicated a large amount of air defense and EM jamming resources to protect the bridge, which increases the difficulty of pulling something off for Ukraine.
For a long time Ukraine didn't have the types of weapons that would be needed to even attempt it outside of saboteur types of actions. Now they have some precise ordnance like Storm Shadow but even these weapons are not destructive enough to take the bridge down except in large quantities, and those are quantities they seem to have decided are best put towards other targets.
Ukraine has recently seen substantial success in finding better weapons, with drones that can engage in "medium-range" scenarios to close off the so-called land bridge from Russia to Crimea. These weapons have also helped in degrading Russia's own defenses, but with this in mind Ukraine may feel it best to leave the Kerch bridge standing for now to allow Russian occupiers to flee across the bridge back to Russia, since Ukraine has the northern land route through occupied territories under much more effective fire control than at any point since 2022.