the robot as seen in the video makes zero sense, a railing camera on the ceiling or top of the opposite shelf would do the same in a fraction of the time because it doesn't have to be slow to avoid injuring people and will have greater accuracy due the fixed nature of the shelfs spatial relationships, moreover it doesn't have to come with complex navigation software and hardware.

research is fine and all, but this is quite far from something that could have resulted in a production run.