While reading an interesting article on LinkedIN by Scott Nyquist, Senior Partner at McKinsey & Company, the thought occurred to me how batch automation and autonomous vehicle (AV) automation have a lot in common. Normal conditions are no problem for either to automatically control the operation, it is the abnormal conditions that cause the problems.
In batch control it is called “exception logic”. This typically requires much more complex algorithm configuration than the normal logic. Of course, in batch production plants, there is typically an operator to deal with a truly abnormal situation that the exception logic did not anticipate.
No such luck with a completely autonomous vehicle. One can imagine that the sensors can determine if there is a red light up ahead and begin the automatic braking process. One could even imagine that if the traffic lights were not working, but there was a policeman in the road deciding which vehicles must stop and which ones are allowed to pass. The “exception logic” algorithms in the AV will surely have to be able to handle this situation.
In his article, he cites perhaps one of the most onerous problems. What should an AV do in a situation in which the choice is to hurt a bystander, the occupants of the car, or another car? Besides the life and death decision to be made, there is the legal ramifications as to who is at fault.
So can we expect totally autonomous control of chemical processes anytime soon? Perhaps the industry could do this once all the applications, algorithms and analytics are worked out for vehicles. Although cars are complex, they are nowhere nearly as complex as a chemical plant. So, like Scott Nyquist says about AVs, I would not expect autonomous batch control to be available anytime soon.
Scott Nyquist Article: Why driverless cars might not hit the road so fast