DETAILED ACTION
Claims 1-20 are presented for examination.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Drawings
The drawings received on 31 August 2022 are accepted.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by US patent 12,210,349 B2 Redford, et al. [herein “Redford”].
Claim 1 recites “1. A system comprising: one or more processors; and one or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause the system to perform operations.” Redford column 52 lines 17-19 disclose “the computer system comprises one or more processors (computers) which carry out the functionality.” Redford column 60 lines 43-45 disclose “non-transitory computer-readable storage medium for programming one or more computers.”
Claim 1 further recites “comprising: executing a simulation representing a driving scenario of a simulated vehicle controlled by an autonomous vehicle controller including: determining a relative location of a simulated object within the simulation with respect to a location of the simulated vehicle.” Redford column 14 lines 37-41 and 47-49 disclose:
A driving scenario captured in a scenario description language format is a high-level description of a driving scenario. A driving scenario has both a static layout, such as road layout (lanes, markings etc.), buildings, road infrastructure etc. and dynamic elements. ….
Dynamic elements include, for example, positions and movement of actors within the static layout (e.g. vehicles, pedestrians, cyclists etc.).
The dynamic elements (e.g. vehicles, pedestrians, cyclists etc.) along the road are absolute locations of simulated objects within the simulation. Redford column 56 lines 11-17 disclose:
a physical property of at least one external object (e.g. location/distance from the agent, speed/velocity/acceleration relative to the agent etc.)
Position of an external object in a field of view of the agent (e.g. angle from centre of image in the case of a camera)
A relative location from the agent corresponds with a relative location of the simulated object from a simulated vehicle.
Claim 1 further recites “determining, based on the relative location of the simulated object, an adjusted location of the simulated object within the simulation at least in part by shifting the relative location of the simulated object to the adjusted location to correct for a simulated miscalibration of a sensor of the simulated vehicle.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 1 further recites “wherein the adjusted location is based at least in part on a distribution of a plurality of corrected locations corresponding to a plurality of respective miscalibrations of a sensor of the simulated vehicle within a range of miscalibrations for the sensor.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
Each respective confounder corresponds with a respective miscalibration. The plurality of confounders within the additive error model correspond with a plurality of location corrections. The conditional distributions are a distribution of locations.
Claim 1 further recites “and wherein a corrected location of the plurality of corrected locations is determined by shifting a location relative to the sensor associated with the relative location to the corrected location to correct for the respective miscalibration of the sensor.” Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 1 further recites “performing a collision check between the simulated vehicle and the simulated object at the adjusted location.” Redford column 50 lines 59-67 disclose:
autonomous vehicles (AVs) that can navigate by themselves on urban roads. Such vehicles must not only perform complex manoeuvres among people and other vehicles, but they must often do so while guaranteeing stringent constraints on the probability of adverse events occurring, such as collision with these other agents in the environments. In order for an AV to plan safely, it is crucial that it is able to observe its environment accurately and reliably.
Determining a probability of a collision corresponds with a collision check between a given autonomous vehicle and simulated objects of the environment.
Claim 1 further recites “and determining, based at least in part an outcome of the collision check, a safety metric associated with the autonomous vehicle controller.” Redford column 57 lines 3-4 discloses “The behaviour of the agent may be classified as safe or unsafe.” Classifying the behaviour of the agent as safe is a safety metric.
Claim 2 further recites “2. The system of claim 1, wherein the safety metric indicates that the autonomous vehicle controller operates safely for a threshold portion of miscalibrations within the range of miscalibrations for the sensor.” Redford column 57 lines 3-4 discloses “The behaviour of the agent may be classified as safe or unsafe.” Classifying the behaviour of the agent as safe is a safety metric.
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
Each range for respective slices corresponds with a threshold portions. Each distribution within each bin corresponds with a range for the miscalibrations for the respective range of miscalibrations for the respective slice for the sensor.
Claim 2 further recites “the operations further comprising: configuring an autonomous vehicle to use the autonomous vehicle controller, based at least in part on the safety metric.” Redford column 8 lines 42-47 discloses “The planning and prediction system predicts the likely trajectories of other agents in the scene and plans a path through the scene that is safe, legal and comfortable. The control system consumes desired trajectories from the planning and prediction system and outputs control signals for the actuators.” The control system outputting control signals for actuating the trajectories corresponds with configuring an autonomous vehicle to use the control system based at least in part on the path being determined to be safe, legal, and comfortable.
Claim 3 further recites “3. The system of claim 1, wherein the adjusted location is determined at least in part by retrieving an offset or a scaling factor from a miscalibration lookup table based on the relative location of the simulated object, the offset or the scaling factor determined based on a boundary location of the distribution, the boundary encompassing a threshold portion of corrected locations for miscalibrations within the range of miscalibrations for the sensor.” From the above list of alternatives the Examiner is selecting “an offset.”
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The range (distance from detector) of a slice for a bin corresponds with a lookup table based on the relative location of the simulated object. A distance from the detector is a relative location of the simulated object. The confounder distribution within the bin corresponds to a respective offset determined based on the bin. Each respective range corresponds with a boundary for the bin and its distribution. The range for the slice corresponds with a threshold portion. Each distribution within each bin corresponds with a range for the miscalibrations of the sensor.
Claim 4 further recites “4. The system of claim 3, wherein retrieving the one or more the offset or the scaling factor comprises retrieving the scaling factor.” Redford column 32 lines 10-11 disclose “For the scale of the normal distribution, an Inverse Gamma prior is used.” The scale distribution for the bin’s distribution corresponds with a scaling factor.
Claim 4 further recites “and wherein the adjusted location is further determined at least in part by applying the scaling factor to a distance between the relative location of the simulated object and a location of the simulated vehicle.” Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The distance from detector corresponds with a distance between the relative location of the simulated object and the simulated vehicle.
Claim 5 further recites “5. The system of claim 3, wherein the boundary location comprises a closest point to the simulated vehicle on the boundary of the distribution.” Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The distance from detector corresponds with a distance between the relative location of the simulated object and the simulated vehicle. The range for the respective slice identifies boundaries of the ranges including boundary locations based on the distance from detector. A person of ordinary skill in the art would understand a distance from detector to correspond with a closest point of the simulated object to the simulated vehicle.
Claim 6 recites “6. One or more non-transitory computer-readable media storing instructions executable by one or more processors, wherein the instructions, when executed, cause the one or more processors to perform operations.” Redford column 52 lines 17-19 disclose “the computer system comprises one or more processors (computers) which carry out the functionality.” Redford column 60 lines 43-45 disclose “non-transitory computer-readable storage medium for programming one or more computers.”
Claim 6 further recites “comprising: executing a simulation representing a driving scenario of a simulated vehicle controlled by an autonomous vehicle controller, wherein executing the simulation includes: determining a relative location of a simulated object within the simulation with respect to a location of the simulated vehicle.” Redford column 14 lines 37-41 and 47-49 disclose:
A driving scenario captured in a scenario description language format is a high-level description of a driving scenario. A driving scenario has both a static layout, such as road layout (lanes, markings etc.), buildings, road infrastructure etc. and dynamic elements. ….
Dynamic elements include, for example, positions and movement of actors within the static layout (e.g. vehicles, pedestrians, cyclists etc.).
The dynamic elements (e.g. vehicles, pedestrians, cyclists etc.) along the road are absolute locations of simulated objects within the simulation. Redford column 56 lines 11-17 disclose:
a physical property of at least one external object (e.g. location/distance from the agent, speed/velocity/acceleration relative to the agent etc.)
Position of an external object in a field of view of the agent (e.g. angle from centre of image in the case of a camera)
A relative location from the agent corresponds with a relative location of the simulated object from a simulated vehicle.
Claim 6 recites “determining, based on the relative location of the simulated object, an adjusted location of the simulated object within the simulation.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 6 recites “and performing a collision check between the simulated vehicle and the simulated object at the adjusted location.” Redford column 50 lines 59-67 disclose:
autonomous vehicles (AVs) that can navigate by themselves on urban roads. Such vehicles must not only perform complex manoeuvres among people and other vehicles, but they must often do so while guaranteeing stringent constraints on the probability of adverse events occurring, such as collision with these other agents in the environments. In order for an AV to plan safely, it is crucial that it is able to observe its environment accurately and reliably.
Determining a probability of a collision corresponds with a collision check between a given autonomous vehicle and simulated objects of the environment.
Claim 6 recites “and determining, based at least in part an outcome of the collision check, a safety metric associated with the autonomous vehicle controller.” Redford column 57 lines 3-4 discloses “The behaviour of the agent may be classified as safe or unsafe.” Classifying the behaviour of the agent as safe is a safety metric.
Dependent claim 7 is substantially similar to claim 2 above and is rejected for the same reasons.
Claim 8 further recites “8. The one or more non-transitory computer-readable media of claim 6, wherein the determining of the adjusted location is based at least in part on shifting the relative location of the simulated object to the adjusted location.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 8 further recites “based at least in part on a distribution of a plurality of corrected locations corresponding to a plurality of respective miscalibrations of a sensor of the simulated vehicle within a range of miscalibrations for the sensor.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
Each respective confounder corresponds with a respective miscalibration. The plurality of confounders within the additive error model correspond with a plurality of location corrections. The conditional distributions are a distribution of locations.
Claim 9 further recites “9. The one or more non-transitory computer-readable media of claim 8, wherein the adjusted location is determined at least in part by retrieving one or more of an offset or scaling factor from a miscalibration lookup table based on the relative location of the simulated object.” From the above list of alternatives the Examiner is selecting “an offset.”
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The range (distance from detector) of a slice for a bin corresponds with a lookup table based on the relative location of the simulated object. A distance from the detector is a relative location of the simulated object. The confounder distribution within the bin corresponds to a respective offset determined based on the bin. Each respective range corresponds with a boundary for the bin and its distribution. The range for the slice corresponds with a threshold portion. Each distribution within each bin corresponds with a range for the miscalibrations of the sensor.
Dependent claim 10 is substantially similar to claim 4 above and is rejected for the same reasons.
Claim 11 further recites “11. The one or more non-transitory computer-readable media of claim 8, wherein a corrected location of the plurality of corrected locations is determined by shifting a location relative to the sensor associated with the relative location to the corrected location to correct for the respective miscalibration of the sensor.” Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 12 further recites “12. The one or more non-transitory computer-readable media of claim 8, wherein: the adjusted location is further determined at least in part by applying a scaling factor to a distance between the relative location of the simulated object and a location of the simulated vehicle; the scaling factor is determined based on a boundary location of the distribution; the boundary encompasses a threshold portion of corrected locations for respective miscalibrations within the range of miscalibrations for the sensor.” Redford column 32 lines 10-11 disclose “For the scale of the normal distribution, an Inverse Gamma prior is used.” The scale distribution for the bin’s distribution corresponds with a scaling factor.
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The distance from detector corresponds with a distance between the relative location of the simulated object and the simulated vehicle. Each range for each slice corresponds with boundaries on the locations for determining respective distributions for the bin. Each range for respective slices corresponds with a threshold portions. Each distribution within each bin corresponds with a range for the miscalibrations for the respective range of miscalibrations for the respective slice for the sensor.
Claim 12 further recites “and the boundary location is a closest point to the simulated vehicle on the boundary of the distribution.” Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The distance from detector corresponds with a distance between the relative location of the simulated object and the simulated vehicle. The range for the respective slice identifies boundaries of the ranges including boundary locations based on the distance from detector. A person of ordinary skill in the art would understand a distance from detector to correspond with a closest point of the simulated object to the simulated vehicle.
Claim 13 recites “13. A method comprising: executing a simulation representing a driving scenario of a simulated vehicle controlled by an autonomous vehicle controller, wherein executing the simulation includes: determining a relative location of a simulated object within the simulation with respect to a location of the simulated vehicle.” Redford column 14 lines 37-41 and 47-49 disclose:
A driving scenario captured in a scenario description language format is a high-level description of a driving scenario. A driving scenario has both a static layout, such as road layout (lanes, markings etc.), buildings, road infrastructure etc. and dynamic elements. ….
Dynamic elements include, for example, positions and movement of actors within the static layout (e.g. vehicles, pedestrians, cyclists etc.).
The dynamic elements (e.g. vehicles, pedestrians, cyclists etc.) along the road are absolute locations of simulated objects within the simulation. Redford column 56 lines 11-17 disclose:
a physical property of at least one external object (e.g. location/distance from the agent, speed/velocity/acceleration relative to the agent etc.)
Position of an external object in a field of view of the agent (e.g. angle from centre of image in the case of a camera)
A relative location from the agent corresponds with a relative location of the simulated object from a simulated vehicle.
Claim 13 further recites “determining, based on the relative location of the simulated object, an adjusted location of the simulated object within the simulation.” Redford column 21 lines 10-19 disclose “Low Level errors” including “calibration errors.” Calibration errors correspond with respective miscalibrations.
Redford column 56 lines 40-44 disclose “Preferably, the PSPM is applied to the perception ground truth and a set of one or more confounders associated with the simulated scenario. The perception ground truth may be computed for each external object using ray tracing.” Redford column 30 section 4.2.1 lines 44-52 disclose:
4.2.1 Positional Errors
The centre position of dynamic objects detected by the perception stack will be modelled using an additive error model given by
y
k
=
x
k
+
e
k
where
y
k
is the observed position of an object,
x
k
is the ground truth position of that object and
e
k
is an error term
The additive error
e
k
corresponds with a respective shift of the relative location of the simulated object to correct for respective errors, including the above calibration errors.
Claim 13 further recites “and performing a collision check between the simulated vehicle and the simulated object at the adjusted location.” Redford column 50 lines 59-67 disclose:
autonomous vehicles (AVs) that can navigate by themselves on urban roads. Such vehicles must not only perform complex manoeuvres among people and other vehicles, but they must often do so while guaranteeing stringent constraints on the probability of adverse events occurring, such as collision with these other agents in the environments. In order for an AV to plan safely, it is crucial that it is able to observe its environment accurately and reliably.
Determining a probability of a collision corresponds with a collision check between a given autonomous vehicle and simulated objects of the environment.
Claim 13 further recites “and determining, based at least in part an outcome of the collision check, a safety metric associated with the autonomous vehicle controller.” Redford column 57 lines 3-4 discloses “The behaviour of the agent may be classified as safe or unsafe.” Classifying the behaviour of the agent as safe is a safety metric.
Dependent claim 14 is substantially similar to claim 2 above and is rejected for the same reasons.
Dependent claim 15 is substantially similar to claim 8 above and is rejected for the same reasons.
Claim 16 further recites “16. The method of claim 15, wherein the adjusted location is determined at least in part by retrieving one or more of an offset or scaling factor from a miscalibration lookup table based on the relative location of the simulated object.” From the above list of alternatives the Examiner is selecting “an offset.”
Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The range (distance from detector) of a slice for a bin corresponds with a lookup table based on the relative location of the simulated object. A distance from the detector is a relative location of the simulated object. The confounder distribution within the bin corresponds to a respective offset determined based on the bin. Each respective range corresponds with a boundary for the bin and its distribution. The range for the slice corresponds with a threshold portion. Each distribution within each bin corresponds with a range for the miscalibrations of the sensor.
Claim 17 further recites “17. The method of claim 16, wherein retrieving the one or more the offset or the scaling factor comprises retrieving the scaling factor.” Redford column 32 lines 10-11 disclose “For the scale of the normal distribution, an Inverse Gamma prior is used.” The scale distribution for the bin’s distribution corresponds with a scaling factor.
Claim 17 further recites “and wherein the adjusted location is further determined at least in part by applying the scaling factor to a distance between the relative location of the simulated object and a location of the simulated vehicle.” Redford column 31 lines 33-45 disclose:
The conditional distribution modelled by PRISM is expected to have a complicated functional form. This functional form can be approximated by discretising each confounder. In this representation, categorical confounders (such as vehicle type) are mapped to bins. Continuous confounders (such as distance from detector) are sliced into ranges and each range mapped to a bin. The combination of these discretisations is a multidimensional table, for which an input set of confounders maps to a bin. It is assumed that within each bin the variance is homoskedastic, and a distribution with constant parameters can be fitted. Global heteroskedasticity is captured by the different parameters in each bin.
The distance from detector corresponds with a distance between the relative location of the simulated object and the simulated vehicle.
Dependent claim 18 is substantially similar to claim 11 above and is rejected for the same reasons.
Dependent claim 19 is substantially similar to claim 12 above and is rejected for the same reasons.
Claim 20 further recites “20. The method of claim 13, wherein the autonomous vehicle controller performs operations of one or more of a perception system, a prediction system, or a planner system.” From the above list of alternatives the Examiner is selecting “a perception system.”
Redford column 8 lines 29-35 discloses:
A perception system 102 receives sensor readings from the world and outputs a scene representation. A planning and prediction system (denoted separately by reference numerals 104 and 106) takes the scene representation and plans a trajectory through the scene. A control system 108 outputs control signals to the world that will cause the vehicle to follow the trajectory.
Conclusion
Prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20210096571 A1 Modalavalasa; Sai Anurag et al.
teaches
Perception Error Models
US 20210097148 A1 Bagschik; Gerrit et al.
Safety Analysis Framework
US 10496766 B2 Levinson; Jesse Sol et al.
Simulation system and methods for autonomous vehicles;
Perception engine for object identification
US 9720415 B2 Levinson; Jesse Sol et al.
Sensor-based object-detection optimization for autonomous vehicles
Holder, M., et al. “Modeling and Simulation of Radar Sensor Artifacts for Virtual Testing of Autonomous Driving” Tagung Automatisiertes Fahren (2019)
Simulation models of sensors;
Radar artifacts.
Barbier, M., et al. “Validation of Perception and Decision-Making Systems for Autonomous Driving via Statistical Model Checking” IEEE Intelligent Vehicles Symposium (2019)
Statistical model checking for decision-making systems including collision risk estimates.
Perception system.
Elmquist, A. & Negrut, D. “Methods and Models for Simulating Autonomous Vehicle Sensors” IEEE Transactions on Intelligent Vehicles, vol. 5, no. 4 (2020)
Models for simulating autonomous vehicle sensors;
Ray tracing.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jay B Hann whose telephone number is (571)272-3330. The examiner can normally be reached M-F 10am-7pm EDT.
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, Renee Chavez can be reached at (571) 270-1104. 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.
/Jay Hann/Primary Examiner, Art Unit 2186 24 November 2025