Generally speaking, no. But there are classes of problems (ex surgical robotics) where solution optimality (or at least near-optimality) is essential to having a usable and safe system.

In control when you say optimal it usually just means that you solve your problem by defining a cost function which you try to minimize within constraints. For example in motion planning your cost could be distance traveled + time spent + energy used. This cost function would be written as a function of the system state (position, velocity, etc.) and control input (e.g. throttle). The control input that limits the cost function is what you're looking for.

Optimal shouldn't be equated to the absolute best way, because it's always a question of definition.