DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Remarks
This final office action is a response to the reply received on 02/24/2026. Claims 1-3, 5-13, and 15-20 are pending. Claims 4 and 14 have been canceled. Claims 1, 5-7, 11, 15-17, and 20 have been amended.
Response to Arguments
Applicant’s arguments/amendments overcome the previous 102 rejection in view of Djuric et. al. (US 20190049970 A1).
Applicant’s additional arguments with respect to the claims have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Information Disclosure Statement
The information disclosure statement (IDS) dated 01/05/2026 has been annotated and considered.
Claim Objections
Claims 6 and 16 are objected to because of the following informalities: The variables defined in the amendment
μ
h
and
σ
h
are repeated twice, where the second instance should read
μ
r
a
n
d
σ
r
(subscripts)
The claim set is objected to because the equations written in the claims filed on 02/24/2026 are blurry/illegible. The examiner recommends submitting a new copy of the claims.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-3, 7-8, 11-13, 17-18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Bacchus (US 20200167934 A1) in view of Rangesh et. al. (US 20210056713 A1).
Regarding Claim 1, Bacchus discloses:
A vehicle control system provided in a vehicle, comprising a vehicle electronic control unit in communication with a vehicle sensor system, a vehicle actuator system, and a map database, the electronic control unit being programmed to: (See at least ¶0044 via "As shown, the vehicle 10 generally includes a propulsion system 20, a transmission system 22, a steering system 24, a brake system 26, a sensor system 28, an actuator system 30, at least one data storage device 32, at least one controller 34, and a communication system 36." and ¶0035 via "…sensed data can be fused together with map data to identify and track objects in the vicinity of the vehicles…")
identify, based on received sensor data from the vehicle sensor system, a second vehicle in a roadway surrounding the vehicle; (See at least ¶0006 via "receiving, by a processing unit disposed in a vehicle, sensor fusion data related to a plurality of target objects and object tracks about the vehicle")
estimate a velocity and a heading of the second vehicle based on the received sensor data; (See at least ¶0008 via "The process model for the track object comprises…to update spline parameters, ν.sub.t+1=ν.sub.t+ΔTa.sub.t to update a longitudinal speed…to update a lane heading wherein u.sub.n is a target spline parameter at discrete time n, a.sub.n is acceleration, and ϕ.sub.n is heading angle.")
extract, from the map database, road information including information on a lane path for a lane in which the second vehicle is traveling; (See at least ¶0042 via "…vehicle that may include a processor for object tracking, lane-assignment and classification tracking, and lane-assignment and classification of a perception model 100. In general, the mapping data is fused into a perception model (or simply “system”) 100. The system 100 determines the correct position of moving target vehicles to better align with the road geometry using map data…")
estimate a first future position of the second vehicle based on the velocity and the heading, wherein the first future position is estimated using a constant velocity heading model that uses only the velocity and the heading, and that assumes the velocity and the heading of the second vehicle will each remain constant; (See at least ¶0009 via "The path unconstraint hypothesis further includes: generating, a process model, for at least constant velocity for the track object which includes: x.sub.t+1, =x.sub.t+ΔTv.sub.t cos ϕ.sub.t, y.sub.t+1=y.sub.t+ΔTv.sub.t sin ϕ.sub.t, ν.sub.t+1=ν.sub.t, a.sub.t+1=a.sub.t and ϕ.sub.t+1=ϕ.sub.t." **Wherein ϕ.sub.t+1=ϕ.sub.t. indicates a constant heading and ν.sub.t+1=ν.sub.t indicates a constant velocity** as well as ¶0008 via "…update a longitudinal speed…update a lane heading…" **Wherein the path-unconstrained process model uses the velocity and heading to estimate a future position, and corresponds to the constant velocity heading model**)
estimate a second future position of the second vehicle based on the velocity and the lane path of the lane in which the second vehicle is traveling, wherein the second future position is estimated using a lane snapping model that uses a magnitude of the velocity and the lane path, and that assumes the second vehicle will follow the lane path with the magnitude of velocity being constant; and (See at least ¶0057 via "the perception model constrains the object 625 to a Frenet frame (i.e. a longitudinal and lateral position along path 620). The position of the object 625 is given by ƒ.sub.x(u), ƒ.sub.y(u)); of splines parameterized by u (longitudinal distance or arbitrary parameter)" and ¶0060 via "lane constrained models 865 track the object by object vectors using Kalman filters based on each lane constrained hypothesis (equal to the number of splines)", as well as ¶0070 via "the Kalman filters 1030 for the lane constrained hypotheses (i.e. input received from the path constrained model 1015) is as follows: first, the EKFs for the constrained hypotheses use a variation of the constant acceleration (CA) and constant velocity (CV) process models. The position of the target is tracked in the Frenet frame, where the longitudinal position is represented by the parameter u.sub.t of a 2-D parametric spline model for the modelling of the center of the lane, and the lateral position is represented by the signed distance parameter l.sub.t from the lane center (note that this parameter may not necessarily be equal to the lateral position)" **Wherein the lane-path information/splines that utilize the center of the lane and the vehicle's velocity (constant velocity) are used to determine a future position which corresponds to the lane-snapping model)
estimated independently by the lane snapping model (See at least "path constrained model 1015" and "the unconstrained model 1020" via at least ¶0064: "The path constrained model 1015, the unconstrained model 1020 (i.e. constant velocity, acceleration, etc. models), and the stationary model 1025 (i.e. where zero speed is assumed for the tracked object) send path and object data to the Kalman filters 1030. The track states 1045 communicate with the Kalman filters 1030 and send track state data to the hypothesis probability update 1070. Also, the hypothesis probability update 1070 receives data from the classification models 1055 because each hypothesis has a corresponding Naïve Bayes model (i.e. classification model 1055) with a likelihood L.sub.i (x)" and ¶0066 via "…multiple hypotheses {H.sub.i} are created at 1075…Each hypothesis has corresponding Naïve Bayes model with likelihood L.sub.i(x)…" **Wherein, Bacchus is independently estimating the future position using at least two separate models, and computing which is most likely).
However, Bacchus does not explicitly disclose combining the first and second future position estimates to estimate a third future position.
Nevertheless, Rangesh--who is directed towards surround vehicle tracking and motion prediction--discloses: estimate a third future position of the second vehicle by combining (See at least ¶0089 via "The trajectory prediction module 1150 outputs a linear combination of the trajectories predicted by a motion model that leverages the estimated instantaneous motion of the surround vehicles and a probabilistic trajectory prediction model which learns motion patterns of vehicles on freeways from a freeway trajectory training set." and ¶0157 via "The future trajectories may be determined by, for example, averaging predicted future locations and covariances provided by the motion model and/or the probabilistic model.").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus' multiple hypotheses in view of Rangesh's concept of combining trajectory predictions from different models in order to improve the prediction accuracy, robustness, and safety of the vehicle control by utilizing linear combination of multiple models and averaging to determine a future trajectory/positions rather than relying only on a single prediction estimate: "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
Regarding Claim 11, Bacchus discloses:
A method for controlling a vehicle, comprising using a vehicle electronic control unit to: (See at least ¶0051 via "The instructions, when executed by the processor 44, receive and process signals (e.g., sensor data) from the sensor system 28, perform logic, calculations, methods and/or algorithms for automatically controlling the components of the vehicle 10, and generate control signals that are transmitted to the actuator system 30 to automatically control the components of the vehicle 10 based on the logic, calculations, methods, and/or algorithms").
(Regarding the method steps, see Claim 1 rejection).
Regarding Claim 20, Bacchus discloses:
A vehicle, comprising: (See at least Figure 1 via vehicle 10)
a vehicle sensor system; (See at least Figure 1 via 'a sensor system 28')
a vehicle actuator system; and (See at least Figure 1 via 'an actuator system 30')
(Regarding the steps, see Claim 1 rejection).
Regarding Claims 2 and 12 respectively, Modified Bacchus discloses the vehicle control system according to Claim 1 and the method for controlling the vehicle according to Claim 11.
Furthermore, Rangesh discloses the "third future position" and further discloses: wherein the vehicle electronic control unit is further programmed to determine a trajectory of the vehicle based on the third future position of the second vehicle (See at least ¶0159 via "The expected future trajectory may be used for a variety of applications. For example, an output representative of the expected future trajectory may be generated. The output may be a control signal responsive to the expected future trajectory and may serve to provide an update and/or a warning to one or more uses who may, for example, be monitoring activity of the vehicle 100. The control signal may be a signal received within the vehicle 100, for example if the vehicle 100 is being operated by a user. The control signal may also be used to adjust the path of the vehicle 100 to, for example, make adjustments in view of the expected future trajectory. This adjustment may be in for example a semi-autonomous or autonomous vehicle.").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus in view of Rangesh's third future position of a second/surrounding vehicle used to control the trajectory of the ego vehicle in order to utilize more accurate/robust future position(s) of a surrounding vehicle "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
Regarding Claims 3 and 13 respectively, Modified Bacchus discloses the vehicle control system according to Claim 2 and the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses a vehicle actuator system (See at least Figure 1 via 'an actuator system 30'), and Rangesh discloses the "third future position" and further: wherein the vehicle electronic control unit is further programmed to control the vehicle actuator system based on the trajectory determined based on the third future position of the second vehicle (See at least ¶0159 via "The expected future trajectory may be used for a variety of applications. For example, an output representative of the expected future trajectory may be generated. The output may be a control signal responsive to the expected future trajectory and may serve to provide an update and/or a warning to one or more uses who may, for example, be monitoring activity of the vehicle 100. The control signal may be a signal received within the vehicle 100, for example if the vehicle 100 is being operated by a user. The control signal may also be used to adjust the path of the vehicle 100 to, for example, make adjustments in view of the expected future trajectory. This adjustment may be in for example a semi-autonomous or autonomous vehicle." **Wherein adjusting the vehicle's path includes controlling an actuator system).
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus in view of Rangesh's third future position of a second/surrounding vehicle used to control the trajectory of the ego vehicle in order to utilize more accurate/robust future position(s) of a surrounding vehicle "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
Regarding Claims 7 and 17 respectively, Modified Bacchus discloses the vehicle control system according to Claim 1 and the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses: wherein the electronic control unit is further programmed to iteratively update the (See at least ¶0037 via " Each instance a sensor fusion containing an object message is received, the corresponding bank of filters is updated along with the probability of each hypothesis"
identifying the second vehicle in the roadway surrounding the vehicle; (See at least ¶0006 via "receiving, by a processing unit disposed in a vehicle, sensor fusion data related to a plurality of target objects and object tracks about the vehicle")
estimating the velocity and the heading of the second vehicle based on the received sensor data; (See at least ¶0008 via "The process model for the track object comprises…to update spline parameters, ν.sub.t+1=ν.sub.t+ΔTa.sub.t to update a longitudinal speed…to update a lane heading wherein u.sub.n is a target spline parameter at discrete time n, a.sub.n is acceleration, and ϕ.sub.n is heading angle.")
extracting road information including information on the lane path for the lane in which the second vehicle is traveling; (See at least ¶0042 via "…vehicle that may include a processor for object tracking, lane-assignment and classification tracking, and lane-assignment and classification of a perception model 100. In general, the mapping data is fused into a perception model (or simply “system”) 100. The system 100 determines the correct position of moving target vehicles to better align with the road geometry using map data…")
estimating the first future position of the second vehicle using the constant velocity heading model; (See at least ¶0009 via "The path unconstraint hypothesis further includes: generating, a process model, for at least constant velocity for the track object which includes: x.sub.t+1, =x.sub.t+ΔTv.sub.t cos ϕ.sub.t, y.sub.t+1=y.sub.t+ΔTv.sub.t sin ϕ.sub.t, ν.sub.t+1=ν.sub.t, a.sub.t+1=a.sub.t and ϕ.sub.t+1=ϕ.sub.t." **Wherein ϕ.sub.t+1=ϕ.sub.t. indicates a constant heading and ν.sub.t+1=ν.sub.t indicates a constant velocity** as well as ¶0008 via "…update a longitudinal speed…update a lane heading…" **Wherein the path-unconstrained process model uses the velocity and heading to estimate a future position, and corresponds to the constant velocity heading model**)
estimating the second future position of the second vehicle using the lane snapping model; and (See at least ¶0057 via "the perception model constrains the object 625 to a Frenet frame (i.e. a longitudinal and lateral position along path 620). The position of the object 625 is given by ƒ.sub.x(u), ƒ.sub.y(u)); of splines parameterized by u (longitudinal distance or arbitrary parameter)" and ¶0060 via "lane constrained models 865 track the object by object vectors using Kalman filters based on each lane constrained hypothesis (equal to the number of splines)", as well as ¶0070 via "the Kalman filters 1030 for the lane constrained hypotheses (i.e. input received from the path constrained model 1015) is as follows: first, the EKFs for the constrained hypotheses use a variation of the constant acceleration (CA) and constant velocity (CV) process models. The position of the target is tracked in the Frenet frame, where the longitudinal position is represented by the parameter u.sub.t of a 2-D parametric spline model for the modelling of the center of the lane, and the lateral position is represented by the signed distance parameter l.sub.t from the lane center (note that this parameter may not necessarily be equal to the lateral position)" **Wherein the lane-path information/splines that utilize the center of the lane and the vehicle's velocity (constant velocity) are used to determine a future position which corresponds to the lane-snapping model).
However, Bacchus does not explicitly disclose the third future position, or the combination of the first and second future positions to estimate the third future position.
Nevertheless, Rangesh discloses: estimating the third future position of the second vehicle by combining the first future position and the second future position (See at least ¶0089 via "The trajectory prediction module 1150 outputs a linear combination of the trajectories predicted by a motion model that leverages the estimated instantaneous motion of the surround vehicles and a probabilistic trajectory prediction model which learns motion patterns of vehicles on freeways from a freeway trajectory training set." and ¶0157 via "The future trajectories may be determined by, for example, averaging predicted future locations and covariances provided by the motion model and/or the probabilistic model.").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus' iterative updating of the vehicle state estimates that are used to form the multiple hypotheses in view of Rangesh's combining trajectory predictions from different models in order to improve the prediction accuracy, robustness, and safety of the vehicle control by utilizing a most updated linear combination of multiple models and averaging to determine a future trajectory/positions over time as new data for the first and second future positions are acquired and updated, rather than relying only on a single prediction estimate: "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
Regarding Claims 8 and 18 respectively, Modified Bacchus discloses the vehicle control system according to Claim 1 and the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses: wherein the electronic control unit is programmed to estimate(See at least ¶0006 via "In one embodiment, a method for enhanced object tracking is provided. The method includes: receiving, by a processing unit disposed in a vehicle, sensor fusion data related to a plurality of target objects and object tracks about the vehicle; determining, by the processing unit, one or more splines representing trajectories of each target object to an object track…")
However, Bacchus does not explicitly disclose the third future position.
Nevertheless, Rangesh discloses: the third future position (See at least ¶0089 via "The trajectory prediction module 1150 outputs a linear combination of the trajectories predicted by a motion model that leverages the estimated instantaneous motion of the surround vehicles and a probabilistic trajectory prediction model which learns motion patterns of vehicles on freeways from a freeway trajectory training set." and ¶0157 via "The future trajectories may be determined by, for example, averaging predicted future locations and covariances provided by the motion model and/or the probabilistic model.").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus' multiple hypotheses/position estimates for a plurality of target vehicles in view of Rangesh's combining trajectory predictions from different models in order to improve the prediction accuracy, robustness, and safety of the vehicle control in an environment with multiple target vehicles/obstacles, by utilizing linear combination of multiple models and averaging to determine future trajectories/positions rather than relying only on a single prediction estimate for each target: "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Bacchus (US 20200167934 A1) and Rangesh et. al. (US 20210056713 A1) in view of Russell et. al. (US 20020049539 A1) and Kodaira (US 20150329108 A1).
Regarding Claims 5 and 15 respectively, Modified Bacchus discloses the vehicle control system according to Claim 1 and the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses: wherein the vehicle electronic control unit is further programmed to: estimate the second future position of the second vehicle assuming the magnitude of velocity will remain constant; and (See at least ¶0057 via "the perception model constrains the object 625 to a Frenet frame (i.e. a longitudinal and lateral position along path 620). The position of the object 625 is given by ƒ.sub.x(u), ƒ.sub.y(u)); of splines parameterized by u (longitudinal distance or arbitrary parameter)" and ¶0060 via "lane constrained models 865 track the object by object vectors using Kalman filters based on each lane constrained hypothesis (equal to the number of splines)", as well as ¶0070 via "the Kalman filters 1030 for the lane constrained hypotheses (i.e. input received from the path constrained model 1015) is as follows: first, the EKFs for the constrained hypotheses use a variation of the constant acceleration (CA) and constant velocity (CV) process models. The position of the target is tracked in the Frenet frame, where the longitudinal position is represented by the parameter u.sub.t of a 2-D parametric spline model for the modelling of the center of the lane, and the lateral position is represented by the signed distance parameter l.sub.t from the lane center (note that this parameter may not necessarily be equal to the lateral position)" **Wherein the lane-path information/splines that utilize the center of the lane and the vehicle's velocity (constant velocity) are used to determine a future position which corresponds to the lane-snapping model)
in estimating, the first future position of the second vehicle (See at least ¶0009 via "The path unconstraint hypothesis further includes: generating, a process model, for at least constant velocity for the track object which includes: x.sub.t+1, =x.sub.t+ΔTv.sub.t cos ϕ.sub.t, y.sub.t+1=y.sub.t+ΔTv.sub.t sin ϕ.sub.t, ν.sub.t+1=ν.sub.t, a.sub.t+1=a.sub.t and ϕ.sub.t+1=ϕ.sub.t." **Wherein ϕ.sub.t+1=ϕ.sub.t. indicates a constant heading and ν.sub.t+1=ν.sub.t indicates a constant velocity** as well as ¶0008 via "…update a longitudinal speed…update a lane heading…" **Wherein the path-unconstrained process model uses the velocity and heading to estimate a future position, and corresponds to the constant velocity heading model**).
However, Bacchus does not explicitly disclose the change in heading and the thresholds.
Nevertheless, Russell--who is directed towards path prediction—discloses: in estimating, (See at least ¶0065 via "The initial step is a process step 340, in which the unfiltered yaw rate measurement of host vehicle 10 is compared to the previous update path heading coefficient multiplied by twice the velocity of host vehicle 10. Decision step 152 then queries whether this value exceeds a predetermined threshold. " *Wherein the yaw rate is the heading over time, and is being tested against a threshold)
if the change in the heading angle of the second vehicle over the predetermined time is not greater than the predetermined threshold, estimate the first future position assuming the velocity and the heading will remain constant, and (See at least ¶0051-0052 via "At decision step 86, the process queries whether the target is on a straight road, not executing a lane change and within an allowable range. If any of these conditions is false, control passes to decision step 90. If all of these conditions are true, control passes to process step 88 where the target positions are propagated forward with longitudinal and lateral target position state vectors and a state transition matrix" *Wherein the vehicle being on a straight road is less than a threshold)
if the change in the heading angle of the second vehicle over the predetermined time is greater than the predetermined threshold, estimate the first future position assuming the velocity will remain constant and that the heading will follow a curve (See at least ¶0065 via "The initial step is a process step 340, in which the unfiltered yaw rate measurement of host vehicle 10 is compared to the previous update path heading coefficient multiplied by twice the velocity of host vehicle 10. Decision step 152 then queries whether this value exceeds a predetermined threshold" as well as ¶0045 via "After filtering the curvature state vector, it is propagated forward with the state transition matrix and the path is propagated forward from the center of the host vehicle in fixed-length steps along an arc." and ¶0053 via "In this process a curve fit is applied to the host vehicle curvature rate data and, for each target, a curve is generated using target history data and data propagated forward from position and velocity estimates in the longitudinal and lateral directions. The curves are both second-order polynomials of the form").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to further modify Modified Bacchus in view of Russell's disclosure of modeling a target vehicle's curvature in order to expand the vehicle scenarios that can be modeled and "to provide a system that is capable of detecting the presence of an obstacle in the forward path of a vehicle when either or both of the vehicle and the obstacle are traveling on either a straight or curved path" [Russell ¶0009].
However, Modified Bacchus does not explicitly disclose the scenario of the decaying to a straight path.
Nevertheless, Kodaira discloses: decaying to a straight path over a predetermined horizon (See at least ¶0080 via "…When the vehicle 2 returns from the gentle curve to the straight road as illustrated in FIG. 8(a), this travels on the curve without adjusting the travel speed before entering the curve, so that it is not required to adjust the travel speed when returning to the straight road (adjust so as to increase the travel speed in FIG. 8(a))…target trajectory set from the gentle curve along the straight road at a constant speed as illustrated in FIG. 8(a).").
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Modified Bacchus in view of the scenario as disclosed by Kodaira in order to account for a broader range of vehicle behavior scenarios by being able to predict how a vehicle would transition from being on a curved path to a straight one, depending on factors such as the curve radius: " the trajectory control is being executed according to the turning radius of the target trajectory in a situation in which the target trajectory is set for the travel from the curve along a straight road at the time of exit from the curve" [Kodaira ¶0079].
Claims 6 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Bacchus (US 20200167934 A1) and Rangesh et. al. (US 20210056713 A1) in view of "Product of Two Gaussian PDFs" (Wayback Machine PDF previously attached in the Non-Final dated 11/26/2025)
Regarding Claims 6 and 16 respectively, Modified Bacchus discloses the vehicle control system according to Claim 1 and the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses: wherein the electronic control unit is programmed to: estimate the first future position by modeling using a Gaussian
N
μ
h
,
σ
h
,
wherein
μ
h
is the mean and
σ
h
is the standard deviation of the first future position; estimate the second future position by modeling using a Gaussian
μ
r
,
σ
r
,
wherein
μ
h
is the mean and
σ
h
is the standard deviation of the second future position; and (See at least ¶0008-¶0009, ¶0057, ¶0060, and ¶0070 via "the Kalman filters 1030 for the lane constrained hypotheses (i.e. input received from the path constrained model 1015) is as follows: first, the EKFs for the constrained hypotheses use a variation of the constant acceleration (CA) and constant velocity (CV) process models…" Additionally, see ¶0066-0068 via "(e.g. L(x)=N(d|μ=100,σ=100) N(ν|μ=0,σ=100)) where N(x|μ,σ) is a Gaussian PDF with a mean μ and a standard deviation σ…")
However, Bacchus does not explicitly disclose the third future position.
Nevertheless, Rangesh discloses: estimate the third future position (See at least ¶0089 via "The trajectory prediction module 1150 outputs a linear combination of the trajectories predicted by a motion model that leverages the estimated instantaneous motion of the surround vehicles and a probabilistic trajectory prediction model which learns motion patterns of vehicles on freeways from a freeway trajectory training set." and ¶0157 via "The future trajectories may be determined by, for example, averaging predicted future locations and covariances provided by the motion model and/or the probabilistic model.")
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Bacchus' multiple hypotheses in view of Rangesh's concept of combining trajectory predictions from different models in order to improve the prediction accuracy, robustness, and safety of the vehicle control by utilizing linear combination of multiple models and averaging to determine a future trajectory/positions rather than relying only on a single prediction estimate: "…safely share the road with human drivers, an autonomous vehicle preferably has the ability to predict the future motion of surrounding vehicles based on perception" [Rangesh ¶0004].
However, modified Bacchus does not explicitly disclose the specific Gaussian equations of
μ
c
o
m
b
i
n
e
d
o
r
σ
c
o
m
b
i
n
e
d
2
as claimed.
Nevertheless, CCRMA Stanford discloses:
N
μ
h
,
σ
h
and the Gaussian
N
μ
r
,
σ
r
; to yield a Gaussian
N
μ
c
o
m
b
i
n
e
d
,
σ
c
o
m
b
i
n
e
d
, where
μ
c
o
m
b
i
n
e
d
=
μ
h
σ
r
2
+
μ
r
σ
h
2
σ
h
2
+
σ
r
2
,
and
σ
c
o
m
b
i
n
e
d
2
=
σ
h
2
σ
r
2
σ
h
2
+
σ
r
2
. (See at least the annotated screen captured figure below which illustrates the equation).
PNG
media_image1.png
341
795
media_image1.png
Greyscale
Modified Bacchus does not explicitly disclose the specific Gaussian mathematical formulas. Nevertheless, applying the mathematical formula of multiplication of two Gaussians to obtain a third such as shown by CCRMA Stanford, would have been an obvious design choice for one of ordinary skill in the art because it facilitates known mathematical means for deriving the probabilistic fusion to obtain the trajectory (See Bacchus ¶0037 via "…hypotheses include different statistical models, which incorporate variables such as the targets coordinates in the Frenet frame (longitudinal and lateral road coordinates) and velocity. Each instance a sensor fusion containing an object message is received, the corresponding bank of filters is updated along with the probability of each hypothesis, the top N hypotheses are outputted as a separate message (where typically N=3). The hypothesis type and probabilities can then be used by downstream modules to determine lane assignment and dynamic properties of objects"). Since the invention did not provide novel or unexpected results from the usage of said claimed formulae, use of any mathematical means, including that of the claimed invention would be an obvious matter of design choice within the skill of the art which would yield predictable results.
PNG
media_image2.png
24
335
media_image2.png
Greyscale
Claims 9-10 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Bacchus (US 20200167934 A1) and Rangesh et. al. (US 20210056713 A1) in view of Djuric et. al. (US 20190049970 A1).
Regarding Claim 9, Modified Bacchus discloses the vehicle control system according to Claim 1.
Furthermore, Bacchus discloses: wherein the road information further includes information on a lane path for at least one additional lane (See at least ¶0062 via "The tracked objects have several splines represented by nearby target trajectories…lane constrained models 965 tracks the object by object vectors using Kalman filters based on each lane constrained hypothesis (equal to the number of splines and coordinate constrained to the spline representation of the lane)…" and ¶0063 via "The candidate paths are all potential trajectories a vehicle may take in relation to a tracked object." as well as Figure 6B. **Wherein the consideration of multiple splines corresponds to the consideration of multiple lanes.)
However, Bacchus does not explicitly state that the lane is adjacent.
Nevertheless, Djuric--who is directed towards object motion prediction and autonomous vehicle control--discloses: at least one additional lane adjacent to the lane in which the second vehicle is traveling (See at least Figure 5 which is an example output of the model (that includes road information) and includes/depicts a second vehicle 202 traveling on a road (travel way 208), and vehicle 302 which is a pedestrian that is traveling on sidewalk 308 and is predicted to travel into bike lane 310 to avoid an obstruction. The sidewalk and bike lane are illustrated as being adjacent to the travel way or lane that the second vehicle is traveling in)
PNG
media_image3.png
688
520
media_image3.png
Greyscale
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Modified Bacchus in view of Djuric's consideration of adjacent lanes, such as those adjacent to where a second vehicle (or pedestrian) is travelling, so that the ego vehicle can consider possible control options/behaviors to best avoid collisions/obstacles with respect to the environment: "The autonomous vehicle can input the combined data set into a machine-learned model to predict how an object is likely to move relative to the characteristics of the geographic area (e.g., how a pedestrian will move relative to a sidewalk with an obstruction, wall, curb, etc.). The machine-learned model can output one or more predicted future locations of the object (e.g., a predicted object trajectory), which the autonomous vehicle can use to plan and control its motion (e.g., to avoid the object)." [Djuric ¶0021].
Regarding Claim 10, Modified Bacchus discloses the vehicle control system according to Claim 9.
Furthermore, Djuric discloses: wherein the at least one additional lane includes one of a crosswalk and a sidewalk, the second vehicle is one of a pedestrian and a bicycle traveling on the one of the crosswalk and the sidewalk (See at least Figure 5 which illustrates at least one additional lane 308 which is a sidewalk, and the second vehicle is a pedestrian traveling on the sidewalk. Additionally see ¶0075).
PNG
media_image4.png
459
346
media_image4.png
Greyscale
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Modified Bacchus in view of Djuric's consideration of adjacent lanes, such as those adjacent to where a second vehicle (or pedestrian) is travelling, so that the ego vehicle can consider possible control options/behaviors to best avoid collisions/obstacles with respect to the environment: "The autonomous vehicle can input the combined data set into a machine-learned model to predict how an object is likely to move relative to the characteristics of the geographic area (e.g., how a pedestrian will move relative to a sidewalk with an obstruction, wall, curb, etc.). The machine-learned model can output one or more predicted future locations of the object (e.g., a predicted object trajectory), which the autonomous vehicle can use to plan and control its motion (e.g., to avoid the object)." [Djuric ¶0021].
Regarding Claim 19, Modified Bacchus discloses the method for controlling the vehicle according to Claim 11.
Furthermore, Bacchus discloses: wherein the road information further includes information on a lane path for at least one additional lane (See at least ¶0062 via "The tracked objects have several splines represented by nearby target trajectories…lane constrained models 965 tracks the object by object vectors using Kalman filters based on each lane constrained hypothesis (equal to the number of splines and coordinate constrained to the spline representation of the lane)…" and ¶0063 via "The candidate paths are all potential trajectories a vehicle may take in relation to a tracked object." as well as Figure 6B. **Wherein the consideration of multiple splines corresponds to the consideration of multiple lanes.)
However, Bacchus does not explicitly state that the lane is adjacent.
Nevertheless, Djuric discloses: at least one additional lane adjacent to the lane...on which the second vehicle is traveling, the at least one additional lane includes one of a crosswalk and a sidewalk, and the second vehicle is one of a pedestrian and a bicycle traveling on the one of the crosswalk and the sidewalk (See at least Figure 5 which is an example output of the model (that includes road information) and includes/depicts a second vehicle 202 traveling on a road (travel way 208), and vehicle 302 which is a pedestrian that is traveling on sidewalk 308 and is predicted to travel into bike lane 310 to avoid an obstruction. The sidewalk and bike lane are illustrated as being adjacent to the travel way or lane that the second vehicle is traveling in)
PNG
media_image3.png
688
520
media_image3.png
Greyscale
Therefore, it would have been obvious to one of ordinary skill in the art prior to the effective filing date of the given invention to modify Modified Bacchus in view of Djuric's consideration of adjacent lanes, such as those adjacent to where a second vehicle (or pedestrian) is travelling, so that the ego vehicle can consider possible control options/behaviors to best avoid collisions/obstacles with respect to the environment: "The autonomous vehicle can input the combined data set into a machine-learned model to predict how an object is likely to move relative to the characteristics of the geographic area (e.g., how a pedestrian will move relative to a sidewalk with an obstruction, wall, curb, etc.). The machine-learned model can output one or more predicted future locations of the object (e.g., a predicted object trajectory), which the autonomous vehicle can use to plan and control its motion (e.g., to avoid the object)." [Djuric ¶0021].
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KAYLA RENEE DOROS whose telephone number is (703)756-1415. The examiner can normally be reached Generally: M-F (8-5) EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abby Lin can be reached on (571) 270-3976. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/K.R.D./Examiner, Art Unit 3657
/ESVINDER SINGH/Examiner, Art Unit 3657
Prosecution Insights