People have always tried to design certain types of robots that resemble humans in some way, and also tried to provide them with expressions as similar to those of humans, with the idea (I suppose) that they would be better accepted by us.
We should not forget the Blade Runner movie in which it was very difficult to differentiate certain robots from humans.
Since then, many different robots have been made with humanoid aspects, and with more or less friendly faces. And many think that if we have a robot in front of us that somehow resembles us, people will react more positively towards them, and interact with them more naturally.
Not long ago we have seen in the media the robot called Nikola that adopted relatively human expressions, and there have been many comments about it.
A video can also be viewed at: https://youtu.be/wfoq7TbJQqs
I am not at all clear that this is necessary for humans to interact with robots in a satisfactory and positive way.
When we have a robot in front of us, and we know that it is, the most important thing is the functions it can perform, and carried in the simplest and most economical way, so that it reaches to the maximum number of people.
Getting good facial expressions is very expensive, it may be fine for a university job, but it does not bring much to those who interact with the robot. Its Cost / Benefit ratio is high.
I understand that researchers carry out work that is at the cutting edge of technology, and that is good, but what ends up being used by the market is what makes economic sense. What you receive based on what you pay for it.
And I think that what the “smilies” provide can be achieved much cheaper with images seen on a screen. Time will tell if I'm right or not.
And in the meantime, what do you think?
EN 
ES