Method for estimating time period until empty for material in a tank of vehicle
During an initialization period of a machine learning model, an electronic data processor is configured to estimate an initial depletion estimate of time period until empty for a material in the tank or container of a machine based on summing initial weighted inputs to the machine learning model in accordance with an initial equation set being applicable to the initialization period that is defined by an initial sub-operation period. After the initialization period of the machine learning model, an electronic data processor is configured to estimate a revised depletion time, where the revised depletion time comprises a time duration until empty or near empty.
1 . A method for estimating a time period until empty or near empty for a material in a tank of a machine, the method comprising:
collecting input data including a consumption rate of the material, a tank capacity for the material, and an initial quantity of the material associated with operation of the machine;
determining a count of sub-operations for the machine, the sub-operations corresponding to respective different fill events for the tank;
when the count is less than a threshold, estimating a first depletion time until empty or near empty for the material in the tank based on a first equation set corresponding to a first sub-operation, wherein near empty is defined as less than a threshold minimum level for the material in the tank;
when the count is greater than or equal to the threshold, estimating a second depletion time until empty or near empty based on a second equation set different from the first equation set, the second equation set corresponding to one or more second sub-operations after the first sub-operation; and
displaying, via an electronic display in communication with an electronic data processor of the machine, at least one of the first depletion time or the second depletion time to an end user of the machine.
2 . The method according to claim 1 , further including determining the first depletion time based on at least one of an observed initial depletion estimate for a respective current or live observation time corresponding to a historical period; a historical estimate of the first depletion time for the historical period; a material type corresponding to the first sub-operation, and an initial consumption rate corresponding to the first sub-operation.
3 . The method according to claim 1 , wherein the first equation set is configured to overweigh an initial depletion estimate determined based on: an observed initial depletion estimate for a current or live observation time corresponding to a historical period; a historical estimate of the first depletion time for the historical period, and a highest ranking available hierarchical factor in a hierarchical succession order.
4 . The method according to claim 3 , wherein the highest ranking available hierarchical factor includes an applied material type and an initially applied material type, the machine includes a sprayer, and the material includes at least one of: a solution, a mixture, a compound, water, nutrients, a fungicide, a pesticide, an insecticide, a herbicide, a biocide, a chemical, or a solvent.
5 . The method according to claim 3 , wherein the highest ranking available hierarchical factor includes an applied material type and an initially applied material type, the machine includes a planter, and the material includes at least one of a seed or a plant.
6 . The method according to claim 1 , wherein the consumption rate represents a respective set of samples accumulated over a sampling interval, and the samples represent an aggregate slope of a predicted decrease in volume of the material in the tank per unit time.
7 . The method according to claim 1 , wherein the consumption rate is based on a machine identifier and one or more available hierarchical factors in hierarchical succession order as follows: (a) a field identifier or work area identifier for a field or work area in which the machine operates, (b) an organization identifier for an organization that owns, leases or operates the machine, and (c) material identifier of the material.
8 . The method according to claim 1 , wherein the initial first equation set includes:
ETEt
=
(
weight
1
2
*
Initial
estimate
t
)
+
(
weight
2
*
Live
sub
‐
operational
estimate
t
)
,
where:
ETEt is the first depletion time until empty or near empty;
weight 1 is a first weight of a first cellular node of a machine learning model;
Initial estimate t is an estimated depletion time until empty or near empty of the material in the tank corresponding to the first sub-operation;
weight 2 is a second weight of a second cellular node of the machine learning model; and
Live sub-operational estimate t is a first observed initial depletion estimate, the first observed initial depletion estimate corresponding to a respective sub-operational period; and
Initial
estimate
t
_
=
MLmodel
(
Live
operational
_
estimate
t
,
HistoricalEstimate
t
,
Materialtype
,
InitialConsumptionrate
,
OptionalModeledFactor
)
,
where:
MLmodel represents the machine learning model;
Live operational estimate t is a second observed initial depletion estimate, the second observed initial depletion estimate corresponding to a current or live observation time, the current or live observation time corresponding to a series of sub-operational periods from a first sub-operational period to a current sub-operational period;
HistoricalEstimate t is a historical estimate of an initial depletion time for a historical period;
Materialtype is a material type of the material;
InitialConsumptionrate includes an initial consumption rate for the material; and
OptionalModeledFactor is an optional factor that is available.
9 . The method according to claim 8 , wherein the machine learning model includes at least one of a gradient boosted tree decision network model, a linear regression model, a decision list network, or a deep learning network.
10 . The method according to claim 8 , wherein the machine learning model includes a network of cellular nodes, wherein an input layer including multiple ones of the cellular nodes is configured to communicate or to propagate an output including the estimated depletion time until empty or near empty.
11 . The method according to claim 1 , wherein the displaying includes:
displaying an estimated remaining coverage area prior to reaching the at least one of the first depletion time or the second depletion time; and
generating an alert message when an observed depletion time is equal to or less than a minimum threshold depletion time.
12 . The method according to claim 1 , wherein the at least one of the first depletion time or the second depletion time includes an estimated time until a fuel tank of the machine is empty, further including controlling or reducing a fuel consumption rate to extend the estimated time.
13 . The method according to claim 1 , further including:
detecting a fill event that adds additional material to existing or remaining material in the tank, wherein the detecting the fill event includes detecting an increase or continuous increase in observed volume of the material in the tank for a threshold number of sample sets and by a threshold increase in volume;
incrementing the count of the sub-operations;
estimating an amount or magnitude of the additional material to be added to the tank; and
determining an updated initial fill level of the tank based on the existing or remaining material and the additional material.
14 . The method according to claim 13 , further including a historical consumption rate in the input data detecting the fill event.
15 . The method according to claim 14 , further including determining the historical consumption rate based on a machine model of the machine, a machine identifier of the machine, an organization identifier of an organization that owns, leases or operates the machine, equipment hours, a field identifier or worksite identifier, a date or season, an ambient temperature, a fuel type, a geographic region in which the machine is operated, and a crop type or operation task identifier.
16 . The method according to claim 1 , wherein the second equation set includes:
Et
E
t
=
(
weight
1
*
Live
operational
estimate
t
)
+
(
weight
2
*
Live
sub
-
operational
estimate
t
)
,
where:
ETE t is the second depletion time until empty or near empty;
weight 1 is a first weight of a first cellular node of a machine learning model, where:
weight
1
=
Tank
Volume
Observed
Max
Tank
Volume
;
Live
operational
estimate
t
=
Tank
Volume
Quantity
op
_
Duration
o
p
,
where:
Tank Volume is a volume of the tank;
Quantity op is a first quantity of the material in the tank based on a continuous series of sub-operational periods from a first sub-operational period to a current sub-operational period;
Duration op is a first duration of the sub-operational periods;
Live operational estimate t is a first observed initial depletion estimate, the first observed initial depletion estimate corresponding to a current or live observation time, the current or live observation time corresponding to the sub-operational periods;
weight
2
=
1
-
weight
1
,
where:
weight 2 is a second weight of a second cellular node of the machine learning model;
and
Live
suboperational
estimate
t
=
Tank
Volume
Quantity
sop
_
Duration
s
o
p
and where:
Quantity sop is a second quantity of the material in the tank for the current sub-operational period;
Duration sop is a second duration of the current sub-operational period; and
Live sub-operational estimate t is a second observed initial depletion estimate corresponding to a respective sub-operational period.
17 . A system comprising:
memory;
instructions in the memory; and
an electronic data processor to execute the instructions to:
collect input data including a consumption rate of material in a tank of a machine, a tank capacity for the material, and an initial quantity of the material associated with operation of the machine;
determine a count of sub-operations for the machine, the sub-operations corresponding to respective different fill events for the tank;
when the count is less than a threshold, estimate a first depletion time until empty or near empty for the material in the tank based on a first equation set corresponding to a first sub-operation, wherein near empty is defined as less than a threshold minimum level for the material in the tank;
when the count is greater than or equal to the threshold, estimate a second depletion time until empty or near empty based on a second equation set different from the first equation set, the second equation set corresponding to one or more second sub-operations after the first sub-operation; and
cause an electronic display to display at least one of the first depletion time or the second depletion time to an end user of the machine.