Define the request without guessing the vehicle
A useful charging request needs a finish time, an availability window, and a representation of the task that the product supports. Some systems know a requested energy amount; others know only a user preference and a charging limit. The prototype must not imply that it knows a vehicle state it cannot read.
The scope documents what comes from the user, what comes from the charging interface, and what is assumed for the study. Those inputs determine which promises the application can reasonably make and which conditions it needs to explain.
Test changes to the plan
A scheduling prototype can compare a normal overnight window with a late arrival, an earlier departure, a manual override, or an unavailable charger. If the request is infeasible, the result should state why rather than silently producing a partial plan.
A controlled study can also examine a household demand constraint or a supplied tariff schedule. Any comparison is tied to the provided inputs and test period. It does not establish a saving for every tariff, vehicle, charger, or household.
Keep instruction, acceptance, and completion separate
Sending a schedule is not evidence that a vehicle has charged. The prototype should keep the proposed plan, device acknowledgement, observed charging state, and task completion distinct. A missing acknowledgement needs its own interface behavior and test case.
The handover can include scheduling code where agreed, a scenario set, input definitions, and a list of information the device interface must supply. This is an illustrative research application. Real equipment access, installation arrangements, and operating responsibilities are agreed before any connected test.