UX for control from multiple places is just really hard:

* How do you know what you have control of

* How do you take/give control

* When you release control, does the other guy's input instantly take over?

* Or does it only take effect when the other guy makes a change something?

* Do you need to be able to see the specific inputs the other guy is giving?

* Do you need to see the actual outputs if the result is actually some merged combination of inputs?

* Under what circumstances should the computer skip scheduled automated tasks due to manual action being taken(See the Gare de Lyon rail disaster)

* Do users need to know about the presence of absence of an automated control input that could, but has not yet taken effect?

* Does anything special need to happen when nobody at all is controlling it?

In home automation, the usual solution is basically just a mutable global, anything can change it, when you're not actively changing it, leave it alone so someone else can... But if you need more than that it gets complicated in a hurry.