import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import com.mongodb.spark._
import com.mongodb.spark.config._
import org.bson.Document
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.types.StructField
import org.apache.spark.sql.types.StringType
import org.apache.spark.sql.types.StructType
import org.apache.spark.sql.Row
import org.apache.spark.sql.types.ArrayType
import org.apache.spark.sql.types.IntegerType
import org.apache.spark.sql.types.DataTypes
import org.apache.spark.sql.types.MapType
import org.apache.spark.sql.types.DateType
import org.apache.spark.sql.types.DoubleType
import org.apache.spark.sql.types.ObjectType
import org.apache.spark.sql.types.DataType
import org.apache.spark.sql.catalyst.expressions.Encode
import org.apache.spark.rdd.RDD

object MongoNestedDocTest {
  def main(args: Array[String]): Unit = {
    
      val sparkSession = SparkSession.builder()
      .master("local")
      .appName("MongoSparkConnector")
      .config("spark.mongodb.input.uri", "mongodb://bigdata:bigdata@servername:27017/bigdata")
      .config("spark.mongodb.output.uri", "mongodb://bigdata:bigdata@servername:27017/bigdata")
      .config("spark.mongodb.output.database", "bigdata")
      .getOrCreate()
      
    val nestedDBRefRDD = sparkSession.sparkContext.parallelize(
      Row("Warsaw", Array(Row("people", Row("100")), Row("people", Row("200")))) ::
      Row("Corted", Array(Row("people", Row("100")), Row("people", Row("200")))) ::
      Nil)
      

    val nestedDBRefSchemaRes = StructType(
      StructField("userName", StringType, false) ::
        StructField("dbRefId", ArrayType(StructType(
            StructField("$ref", StringType, true) ::
            StructField("$id", StructType(StructField("oid", StringType, true) :: Nil), false) ::
            Nil))
            ) :: Nil)

    val nestedDBRefDF = sparkSession.createDataFrame(nestedDBRefRDD, nestedDBRefSchemaRes)
    nestedDBRefDF.show()
    nestedDBRefDF.write.option("collection", "rddMerge").mode("overwrite").format("com.mongodb.spark.sql").save()

  }
}

You will see the following exception:
{color:red}Caused by: com.mongodb.spark.exceptions.MongoTypeConversionException: Cannot cast [100] into a StructType(StructField(oid,StringType,true))
	at com.mongodb.spark.sql.MapFunctions$.com$mongodb$spark$sql$MapFunctions$$convertToBsonValue(MapFunctions.scala:92)
	at com.mongodb.spark.sql.MapFunctions$$anonfun$rowToDocument$1.apply(MapFunctions.scala:59)
	at com.mongodb.spark.sql.MapFunctions$$anonfun$rowToDocument$1.apply(MapFunctions.scala:57)
	at scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:33)
	at scala.collection.mutable.ArrayOps$ofRef.foreach(ArrayOps.scala:186)
	at com.mongodb.spark.sql.MapFunctions$.rowToDocument(MapFunctions.scala:57)
	at com.mongodb.spark.sql.MapFunctions$$anonfun$8.apply(MapFunctions.scala:177){color}

If you change data into 24 digit number or characters such as below, it works fine.

    val nestedDBRefRDD = sparkSession.sparkContext.parallelize(
      Row("Warsaw", Array(Row("people", Row("59b61d2487aa56da25cb9034")), Row("people", Row("599fccd2b9a88e237cb1597c")))) ::
      Row("Corted", Array(Row("people", Row("599fcd76b9a88e237cb1597f")), Row("people", Row("599fccffb9a88e237cb1597d")))) ::
      Nil)