pymongo - Upsert collection to add object with nested objects and nested data - mongodb

Question

I have spent weeks researching, finally asking the question....
Using python/pymongo I am stuck at the simple step of just adding a nested object with nested data "to an existing object or nested object"...so I have not even made it to the upsert yet. I "have" code that confirms if the collection "serial_number+ exists, if not then create it -- however I am stuck at adding nested objects to an existing or newly created collection. Looking for an example to build from, not for this forum to solve the full solution :-)
Attached is a screenshot of the desired layout. Use case, several cars are available to be rented
MongoDB Database with a Collection that is a car's serial number and all activities related to that car
a. Need to capture details of each rental activity - nested data under a each serial number.
b. Need to capture maintenance details for each car - nested data under each serial number.
When either "a" or "b" occur, need to check that the record is not already there, if not then add a new nested object that contains nested data.
I have tried for loops, numerous postings referring to collections and sub-collections -- all seem to require highly detailed information within code to parse the dictionary (which comes from semi-structed sources, not consistent) and add to a collection....but cannot find a starting point to add to an existing object or nested object within an existing collection?
#Assume mongodb is structured like this:
#rent_activity(DB)>vehicle(document)>serial_number(object)>car_details(object under serial_number)
#rent_activity(DB)>vehicle(document)>serial_number(object)>rental_activity(object under serial_number)
#rent_activity(DB)>vehicle(document)>serial_number(object)>maintenance(object under serial_number)
#The example below is attempting to insert new objects with nested arrays to the car_details object
import os,sys
import pymongo
import csv
import boto3
import time
import psycopg2
import pandas as pd
import pandas.io.sql as sqlio
from datetime import datetime
from pymongo import errors
my_connection = psycopg2.connect(dbname='xxx', host='xxx', port=xxx, user='xxx', password='xxx')
client = pymongo.MongoClient("xxxxxxx")
mydb = client["rent_activity"] #This is the mongodb database
collection = mydb["vehicle"]
car_details_sql = "SELECT serial_number, mileage, fuel_level other from car_details"
data = sqlio.read_sql_query(car_details_sql, my_connection) #Create dataframe from query results
try:
#Goal is to add each line from the dataframe result returned from the SQL queary, as a record under collection "vehicle", object "serial_number", nested object "car_details"
for item in data.to_dict(orient="records"): #For each row, transform to dict
mycarkey = str(item.get("purchase_date")) + "_" + str(item.get("serial_number")) #Create unique string
car_number = "car_" + mycarkey
finaldata = {}#Create empty
finaldata[mycarkey] = item #Add outer unique key to existing dict for this set of data
finaldata.update({"_id": car_number}) #Define custom object _id for this set of data
collection.insert_one(finaldata)
#This inserts to the collection "vehicle"
#What is needed is to nest this to:
# rent_activity(DB)>vehicle(document)>serial_number(object)>car_details(object under serial_number)
#NOTE: If I simply add another outer key "serial_number" I receive error key already exists
#..so the issue is how to add an object with nested values beneath an existing object??
except Exception as e:
raise e