Existence and boundedness of optimal controls in infinite-horizon problems