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 .
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are: data processing module, clustering module, comparing module, training module in claim 6.
Because these claim limitations are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, they are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have these limitations interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitations to avoid them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitations recites sufficient structure to perform the claimed function so as to avoid them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112(b)
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1,5-6,10-11,15 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1, 6 recites the limitation "clustering algorithm to group together readings associated with a common road sign into clusters" and later recites “training a machine learning model to correct the location of the readings in each group”. The Claim is indefinite because it is unclear whether “each group” refers to the previously recited “clusters” or to a different set of readings. This inconsistency use of terminology renders the scope of the claim unclear.
Claim 11 recites the limitation " clustering algorithm to group together readings associated with a common road sign into groups" and later recites “train a machine learning model to correct the location of the readings in each cluster”. The Claim is indefinite because it is unclear whether “each cluster” refers to the previously recited “groups” or to a different set of readings. This inconsistency use of terminology renders the scope of the claim unclear.
Claim 5,10,15 recites the limitation "the raw dataset" in the second line. There is insufficient antecedent basis for this limitation in the claim. No raw dataset is previously introduced in claim 1 or claim 5, and therefore it is unclear what dataset is being referenced.
Claim 5,10,15 recites the limitation "running the same clustering method" refers to “clustering algorithm” recited in claim 1 or to a different clustering method. As a result, the scope of the claim cannot be determined with reasonable certainty.
Claim 5,10,15 recites the limitation "the same clustering method to produce the location of signs" the claim is indefinite because it is unclear how the clustering method is used to “produce the location of signs”. In claim 1, the clustering algorithm is used to “group together readings associated with a common road sign into clusters” and no explanation is provided ass to how that clustering operation produce a location. This renders the scope of the claim unclear.
Claim 5,10,15 recites the limitation "from all over" in the second line of the claim. There is insufficient antecedent basis for this limitation in the claim. It is not clear what this is referring to. For the purpose of examination, it is interpreted as the location of all signs in the raw dataset.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 1-15 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
MPEP 2106 (Ill) sets out steps for evaluating whether a claim is drawn to patent-eligible subject matter. The analysis of claims 1-15, in accordance with these steps, follows.
Step 1 Analysis:
Claims 1-5 are directed to method (processes). Claim 6-10 are directed to a device (machine). Claims 11-15 are directed to a computer program product (article of manufacture). Therefore, claims 1-15 fall into one of four statutory categories (i.e., process, machine, article of manufacture).
As to claim 1,
Step 2A Prong 1: this claim recites the following abstract ideas:
using a clustering algorithm to group together readings associated with a common road sign into clusters; (the limitation describes the act of sorting scattered observations into groups based on their similarities to identify which ones belong to the same object, which is a mental process implemented in the human mind. See MPEP 2106.04(a)(2))
comparing the locations of the readings in each cluster to respective tagged road-signs dataset associated with a correct location of the road sign associated with the cluster, to yield a correction vector; and (this limitation describes the comparison of road sign data to determine location is a mental process implemented in the human mind)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
Obtaining an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i).)
training a machine learning model to correct the location of the readings in each group by learning the correction vector. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i)).
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 6,
Step 2A Prong 1: this claim recites the following abstract ideas:
a clustering module configured to use a clustering algorithm to group together readings associated with a common road sign in clusters; (the limitation describes a module that preforms the act of sorting scattered observations into groups based on their similarities to identify which ones belong to the same object, which is a mental process implemented in the human mind. See MPEP 2106.04(a)(2)).
a comparing module configured to compare the locations of each cluster to respective tagged road-signs dataset associated with a correct location, to yield a correction vector; and (this limitation describes the comparison of road sign data to determine location is a mental process implemented in the human mind.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
A data processing module configured to obtain an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i)).
A training module configured to train a machine learning model to correct the location of the readings in each group by learning the correction vectors. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i)).
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 11,
Step 2A Prong 1: this claim recites the following abstract ideas:
Apply a clustering algorithm to group together readings associated with a common road sign into groups; (the limitation describes the act of sorting scattered observations into groups based on their similarities to identify which ones belong to the same object, which is a mental process implemented in the human mind. See MPEP 2106.04(a)(2))
compare the locations of each cluster to respective tagged road-signs dataset associated with a correct location of the road sign associated with the group, to yield a correction vector; and (this limitation describes the comparison of road sign data to determine location is a mental process implemented in the human mind)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
Obtain an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (this limitation describes data collection/receiving, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i).)
Train a machine learning model to correct the location of the readings in each cluster by learning the correction vectors. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer to process, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i)).
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 2,7, and 12
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
Running the machine learning model on the training dataset of readings. (This limitation is directed to mere instruction to apply the abstract idea on a generic computer, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i).)
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 3, 8, and 13
Step 2A Prong 1: this claim recites the following abstract ideas:
The clustering, the comparing, and the training. (These limitations describe repeatedly performing evaluation, and comparison which are mental process involving judgment and decision making, see MPEP 2106.04(a)(2)).
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
Until the machine learning model stops improving (This limitation is directed to mere instruction to apply the abstract idea on a generic computer, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 4, 9, and 14
Step 2A Prong 1: this claim recites the following abstract ideas:
repair the entire raw dataset. (The limitation describes a form of data analysis processing of information. Which is a mental process, see MPEP 2104(a)(2).)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
Using the trained machine learning model to (This limitation is directed to mere instruction to apply the abstract idea on a generic computer, which is a well-understood, routine, conventional activity, see MPEP 2106.05(d)(II)(i))
The additional element does not integrate the judicial exception into practical application and does not amount to significantly more than the Judicial exception.
As to claim 5, 10, and 15
Step 2A Prong 1: this claim recites the following abstract ideas:
Running the same clustering method to produce the location of signs from all over the raw dataset. (The limitation describes organizing and categorizing information to determine locations, which is an evaluation and judgment activity that can be performed as a mental process.)
Step 2A Prong 2 and 2B: the claim recited the following additional elements:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) I.), failing step 2A prong 2. The claim is ineligible.
Claim Rejections - 35 USC § 103
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.
Claim(s) 1-2,5-7,10-12,15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tchuente et al. (Providing more regular road signs infrastructure updates for connected driving: A crowdsourced approach with clustering and confidence level) in view of Chen et al. (US 9459626 B2).
As per claim 1 Tchuente as modified by Chen teaches A method of generating a machine learning model for reducing a location error of readings of road signs sensed by connected vehicle, said method comprising:
Obtaining an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (see [Abstract] “incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs”, and see [2. Related work] “the camera detections of the road signs with attributes such as the latitude and longitude of the detection, the heading of the vehicle, the detection timestamp, the road sign type and the road sign value, for making their daily cloud-based updates of the road sign map.”, and see [3.3. Road sign observations] “Table 1”)
PNG
media_image1.png
814
676
media_image1.png
Greyscale
Using a clustering algorithm to group together readings associated with a common road sign into clusters; (see [Abstract] “tracking road sign infrastructure changes by incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs, while removing noise due to the imprecision of GPS positions in addition to false positive and false negative detections. This goal is achieved by using non-supervised geospatial clustering techniques….”)
Comparing the locations of the readings in each cluster to respective tagged road-signs dataset associated with a correct location of the road sign associated with the cluster, to yield a correction vector; and (see [5.1.2. Data analysis parameters] “The consolidation data processing step described in the previous sections was used to generate the clustered road sign positions with their confidence level.”, and see [5.1.3. Experiment outputs] “The road signs positions and their various confidence levels were compared with real road sign positions collected on the ground.”)
Tchuente does not teach “to yield a correction vector”, and “training a machine learning model to correct the location of the readings in each group by learning the correction vector.”
However, Chen teaches
Using a clustering algorithm to group together readings associated with a common road sign into clusters; (see [Col 4 L 57] the identified static object/road sign within the bounding box is combined or merged with other similarly identified objects/road signs within the bounding box. The merger is performed because one sign may have multiple instances of detection by the vehicle sensor.)
To yield a correction vector; and (see [Col 4 L 64] “in order to train the processor's identification the presence or absence of a road sign (or a type of road sign), the identifications may be verified with ground truth data”, and see [Col 5 L 4] “ an observation of a static object within 30 meters of the ground truth position is defined as a positive example of an accurate detection of a road sign, while an identification of a road sign beyond 100 meters of the ground truth position is defined as a negative example.”, and see [Col 6 L 17] “In one example, a ground truth data set of 148 fixed road signs and 32 variable road signs were placed along a road segment. Using vehicle camera sensors and the machine learning algorithm, the physical location of the road signs were determined. The table below depicts the placement accuracy of the identified signs. For example, 91% of the fixed road sign locations were calculated within 30 meters of their actual location, 81% of the fixed road sign locations were determined to be within 15 meters of their actual location, and so on. Additionally, 97% of the variable road sign locations were determined to be within 30 meters of their actual location, 94% of the variable road sign locations were determined to be within 30 meters of their actual location, and so on”, and see [Col 6 L 39] “In some embodiments, the road sign identification and location calculations may be compared with alternative approaches, such as a clustering-based approach for road sign identification. The comparisons may provide that this machine learning based approach detects less false positives and provides a more robust analysis under various environ mental conditions.” and see table)
PNG
media_image2.png
92
318
media_image2.png
Greyscale
Training a machine learning model to correct the location of the readings in each group by learning the correction vector. (See [Col 5 L 48] “Once a road sign or other targeted static object is identified, the location or placement of the sign/object is determined. In certain embodiments, a machine learning algorithm is used to determine the location of the sign. A linear regression analysis may be used by the machine learning algorithm to place the sign.”, and see [Col 5 L 54] “vehicle probe metadata, such as the speed of the vehicle and the location of the vehicle, are used in the linear regression analysis to place the location of the sign … multiple observation points from the vehicle sensor are compiled and used to place the location of the sign.”)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the invention of Tchuente to include a machine learning model and a correction vector to determines the geographic location of each identified road sign. [Col 8 L 11]
As per claim 2, Tchuente as modified by Chen teaches the method of claim 1.
Tchuente does not teach “further comprising running the machine learning model on the training dataset of readings.”
However, Chen teaches further comprising running the machine learning model on the training dataset of readings. (See [Col 6 L 17] “a ground truth data set of 148 fixed road signs and 32 variable road signs were placed along a road segment. Using vehicle camera sensors and the machine learning algorithm, the physical location of the road signs were determined.”)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the invention of Tchuente to run the machine learning model on the training dataset to determine the geographic location of each road sign with ground truth data. [Col 8 L 15]
As per claim 5, Chen teaches further comprising running the same clustering method to produce the location of signs from all over the raw dataset. (See Chen [Col 4 L 57] “road sign within the bounding box is combined or merged with other similarly identified objects/road signs within the bounding box”, and see Chen [Col 6 L 47] “Following identification of the road signs and determination of the road sign locations, the results may be stored or published within the map developer database.”)
As per claim 6, Tchuente as modified by Chen teaches A system for generating a machine learning model for reducing a location error of readings of road signs sensed by connected vehicle, said system comprising:
obtain an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (see [Abstract] “incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs”, and see [2. Related work] “the camera detections of the road signs with attributes such as the latitude and longitude of the detection, the heading of the vehicle, the detection timestamp, the road sign type and the road sign value, for making their daily cloud-based updates of the road sign map.”, and see [3.3. Road sign observations] “Table 1”)
PNG
media_image1.png
814
676
media_image1.png
Greyscale
use a clustering algorithm to group together readings associated with a common road sign in clusters; (see [Abstract] “tracking road sign infrastructure changes by incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs, while removing noise due to the imprecision of GPS positions in addition to false positive and false negative detections. This goal is achieved by using non-supervised geospatial clustering techniques….”)
compare the locations of each cluster to respective tagged road-signs dataset associated with a correct location; and (see [5.1.2. Data analysis parameters] “The consolidation data processing step described in the previous sections was used to generate the clustered road sign positions with their confidence level.”, and see [5.1.3. Experiment outputs] “The road signs positions and their various confidence levels were compared with real road sign positions collected on the ground.”)
Tchuente does not teach “a data processing module configured to”, “a clustering module configured to”, “a comparing module configured to”, “to yield a correction vector”, and “a training module configured to train a machine learning model to correct the location of the readings in each group by learning the correction vectors.”
However, Chen teaches
a data processing module configured to; (see [Col 1 L 48] “the apparatus comprises at least one processor and at least one memory including computer program code for one or more programs.”)
a clustering module configured to use a clustering algorithm to group together readings associated with a common road sign in clusters; (see [Col 1 L 48] “the apparatus comprises at least one processor and at least one memory including computer program code for one or more programs”, and see [Col 4 L 57] the identified static object/road sign within the bounding box is combined or merged with other similarly identified objects/road signs within the bounding box. The merger is performed because one sign may have multiple instances of detection by the vehicle sensor.)
a comparing module configured to; (see [Col 1 L 48] “the apparatus comprises at least one processor and at least one memory including computer program code for one or more programs.”)
to yield a correction vector; and (see [Col 4 L 64] “in order to train the processor's identification the presence or absence of a road sign (or a type of road sign), the identifications may be verified with ground truth data”, and see [Col 5 L 4] “ an observation of a static object within 30 meters of the ground truth position is defined as a positive example of an accurate detection of a road sign, while an identification of a road sign beyond 100 meters of the ground truth position is defined as a negative example.”, and see [Col 6 L 17] “In one example, a ground truth data set of 148 fixed road signs and 32 variable road signs were placed along a road segment. Using vehicle camera sensors and the machine learning algorithm, the physical location of the road signs were determined. The table below depicts the placement accuracy of the identified signs. For example, 91% of the fixed road sign locations were calculated within 30 meters of their actual location, 81% of the fixed road sign locations were determined to be within 15 meters of their actual location, and so on. Additionally, 97% of the variable road sign locations were determined to be within 30 meters of their actual location, 94% of the variable road sign locations were determined to be within 30 meters of their actual location, and so on”, and see [Col 6 L 39] “In some embodiments, the road sign identification and location calculations may be compared with alternative approaches, such as a clustering-based approach for road sign identification. The comparisons may provide that this machine learning based approach detects less false positives and provides a more robust analysis under various environ mental conditions.” and see table)
PNG
media_image2.png
92
318
media_image2.png
Greyscale
a training module configured to train a machine learning model to correct the location of the readings in each group by learning the correction vectors. . (See [Col 5 L 48] “Once a road sign or other targeted static object is identified, the location or placement of the sign/object is determined. In certain embodiments, a machine learning algorithm is used to determine the location of the sign. A linear regression analysis may be used by the machine learning algorithm to place the sign.”, and see [Col 5 L 54] “vehicle probe metadata, such as the speed of the vehicle and the location of the vehicle, are used in the linear regression analysis to place the location of the sign … multiple observation points from the vehicle sensor are compiled and used to place the location of the sign.”)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the invention of Tchuente to include a machine learning model and a correction vector to determines the geographic location of each identified road sign. [Col 8 L 11]
As per claim 7, this is directed to a system or a computing device claim that corresponds to method claim 2. See the rejection for claim 2 above, which also applies to claim 7.
As per claim 10, this is directed to a system or a computing device claim that corresponds to method claim 5. See the rejection for claim 5 above, which also applies to claim 10.
As per claim 11, Tchuente as modified by Chen teaches
obtain an incoming stream of readings of road signs captured by connected vehicles traveling along roads, wherein each reading contains a location of a road sign and road sign metadata associated therewith; (see [Abstract] “incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs”, and see [2. Related work] “the camera detections of the road signs with attributes such as the latitude and longitude of the detection, the heading of the vehicle, the detection timestamp, the road sign type and the road sign value, for making their daily cloud-based updates of the road sign map.”, and see [3.3. Road sign observations] “Table 1”)
PNG
media_image1.png
814
676
media_image1.png
Greyscale
apply a clustering algorithm to group together readings associated with a common road sign into groups; (see [Abstract] “tracking road sign infrastructure changes by incrementally consolidating crowdsourced cameras' detections of road signs and computing the real positions of the signs, while removing noise due to the imprecision of GPS positions in addition to false positive and false negative detections. This goal is achieved by using non-supervised geospatial clustering techniques….”)
compare the locations of each cluster to respective tagged road-signs dataset associated with a correct location of the road sign associated with the group; (see [5.1.2. Data analysis parameters] “The consolidation data processing step described in the previous sections was used to generate the clustered road sign positions with their confidence level.”, and see [5.1.3. Experiment outputs] “The road signs positions and their various confidence levels were compared with real road sign positions collected on the ground.”)
Tchuente does not teach “A non-transitory computer readable medium for generating a machine learning model for reducing a location error of readings of road signs sensed by connected vehicles, the computer readable medium comprising a set of instructions that when executed cause at least one computer processor to:”, “to yield a correction vector”, and “and train a machine learning model to correct the location of the readings in each cluster by learning the correction vectors.”
However, Chen teaches
A non-transitory computer readable medium for generating a machine learning model for reducing a location error of readings of road signs sensed by connected vehicles, the computer readable medium comprising a set of instructions that when executed cause at least one computer processor to: (see [Col 1 L 48] “the apparatus comprises at least one processor and at least one memory including computer program code for one or more programs.”)
apply a clustering algorithm to group together readings associated with a common road sign into groups; (see [Col 4 L 57] the identified static object/road sign within the bounding box is combined or merged with other similarly identified objects/road signs within the bounding box. The merger is performed because one sign may have multiple instances of detection by the vehicle sensor.)
To yield a correction vector; and (see [Col 4 L 64] “in order to train the processor's identification the presence or absence of a road sign (or a type of road sign), the identifications may be verified with ground truth data”, and see [Col 5 L 4] “ an observation of a static object within 30 meters of the ground truth position is defined as a positive example of an accurate detection of a road sign, while an identification of a road sign beyond 100 meters of the ground truth position is defined as a negative example.”, and see [Col 6 L 17] “In one example, a ground truth data set of 148 fixed road signs and 32 variable road signs were placed along a road segment. Using vehicle camera sensors and the machine learning algorithm, the physical location of the road signs were determined. The table below depicts the placement accuracy of the identified signs. For example, 91% of the fixed road sign locations were calculated within 30 meters of their actual location, 81% of the fixed road sign locations were determined to be within 15 meters of their actual location, and so on. Additionally, 97% of the variable road sign locations were determined to be within 30 meters of their actual location, 94% of the variable road sign locations were determined to be within 30 meters of their actual location, and so on”, and see [Col 6 L 39] “In some embodiments, the road sign identification and location calculations may be compared with alternative approaches, such as a clustering-based approach for road sign identification. The comparisons may provide that this machine learning based approach detects less false positives and provides a more robust analysis under various environ mental conditions.” and see table)
PNG
media_image2.png
92
318
media_image2.png
Greyscale
and train a machine learning model to correct the location of the readings in each cluster by learning the correction vectors. (See [Col 5 L 48] “Once a road sign or other targeted static object is identified, the location or placement of the sign/object is determined. In certain embodiments, a machine learning algorithm is used to determine the location of the sign. A linear regression analysis may be used by the machine learning algorithm to place the sign.”, and see [Col 5 L 54] “vehicle probe metadata, such as the speed of the vehicle and the location of the vehicle, are used in the linear regression analysis to place the location of the sign … multiple observation points from the vehicle sensor are compiled and used to place the location of the sign.”)
As to claim 12, this is directed to a computer-program embodiment that corresponds to method claim 2. See the rejection for claim 2 above, which also applies to claim 12.
As to claim 15, this is directed to a computer-program embodiment that corresponds to method claim 5. See the rejection for claim 5 above, which also applies to claim 15.
Claim(s) 3,8,13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tchuente et al. (Providing more regular road signs infrastructure updates for connected driving: A crowdsourced approach with clustering and confidence level) in view of Chen et al. (US 9459626 B2) and Lin et al. (A general iterative clustering algorithm).
As per claim 3, Tchuente as modified by Chen teaches the method of claim 1.
Tchuente - Chen does not teach “further comprising repeating: the clustering, the comparing, and the training until the machine learning model stops improving.”
However, Lin teaches further comprising repeating: the clustering, the comparing, and the training until the machine learning model stops improving. (see [Abstract] “We introduce a general iterative cluster (GIC) algorithm that improves the proximity matrix and clusters of the base RF. The cluster labels are used to grow a new RF yielding an updated proximity matrix, which is entered into the clustering program. The process is repeated until convergence.”, and see [2.2 The general iterative cluster algorithm] “The GIC algorithm begins by running the underlying or base classification method using an initialization procedure as required to obtain a proximity matrix, followed by running the selected cluster algorithm. Thereafter, each iteration uses the same base classifier followed by the same clustering algorithm. At each step, units are labeled according to the cluster to which they were assigned in the cluster algorithm in the previous iteration. The process continues until convergence. Convergence occurs when the assignment of units to clusters does not change, which corresponds to a proximity matrix that does not change.”)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the invention of Tchuente – Chen to repeat the clustering, comparing, and training until it stops improving to improve the model.
As per claim 8, this is directed to a system or a computing device claim that corresponds to method claim 3. See the rejection for claim 3 above, which also applies to claim 8.
As to claim 13, this is directed to a computer-program embodiment that corresponds to method claim 3. See the rejection for claim 3 above, which also applies to claim 13.
Claim(s) 4,9,14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tchuente et al. (Providing more regular road signs infrastructure updates for connected driving: A crowdsourced approach with clustering and confidence level) in view of Chen et al. (US 9459626 B2) and Krishnan et al. (ActiveClean: interactive data cleaning for statistical modeling).
As per claim 4, Tchuente as modified by Chen teaches the method of claim 1.
Tchuente does not teach “further comprising using the trained machine learning model to repair the entire raw dataset.”
However, Chen teaches “raw dataset” (See [Col 6 L 47] “Following identification of the road signs and determination of the road sign locations, the results may be stored or published within the map developer database.”)
Chen does not teach “further comprising using the trained machine learning model to repair the entire”
However, Krishnan teaches further comprising using the trained machine learning model to repair the entire raw dataset. (see [2.2 Iteration in Model Construction] “The straight-forward application of data cleaning is to repair the corruption in-place, and re-train the model after each repair”)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the invention to modify the invention of Tchuente to use the trained machine learning model to repair the entire raw dataset in order to achieve a better optimization and accuracy.
As per claim 9, this is directed to a system or a computing device claim that corresponds to method claim 4. See the rejection for claim 4 above, which also applies to claim 9.
As to claim 14, this is directed to a computer-program embodiment that corresponds to method claim 4. See the rejection for claim 4 above, which also applies to claim 14.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABDULLAH K ABOUD whose telephone number is (571)272-0025. The examiner can normally be reached Mon-Fri 8am-5pm.
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, Li B Zhen, can be reached at (571) 272-3768. 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.
/ABDULLAH KHALED ABOUD/ Examiner, Art Unit 2121
/Li B. Zhen/ Supervisory Patent Examiner, Art Unit 2121