Patent Issued for Safe hand-off between human driver and autonomous driving system (USPTO 11940790): Allstate Insurance Company

2024 APR 12 (NewsRx) -- By a News Reporter-Staff News Editor at Insurance Daily News -- A patent by the inventors Aragon, Juan Carlos (Redwood City, CA, US), Madigan, Regina (Mountain View, CA, US), filed on December 11, 2019, was published online on March 26, 2024, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents.

Patent number 11940790 is assigned to Allstate Insurance Company (Northbrook, Illinois, United States).

The following quote was obtained by the news editors from the background information supplied by the inventors: “Some automobiles now provide a feature or mode wherein the automobile can drive itself, in some scenarios. This mode may be referred to as an autonomous driving mode, and may be implemented in an autonomous driving system in the vehicle. In many driving situations, the autonomous driving mode may be safer than a non-autonomous mode, in which a human driver controls the operation of the vehicle. However, there will be driving situations in which an unexpected event occurs, and the autonomous mode may not be prepared to handle the event gracefully or safely. In some such cases, the autonomous driving system, being unprepared for the situation, may resort to a default behavior, such as pulling over and stopping. However, in a least some of these cases, a human driver may have been able to safely handle the driving situation, and it would not have been necessary to pull over and stop. Accordingly, there is a need for a method to determine when a human driver can be expected to safely take control of a vehicle, in response to an unexpected event, and to cause an autonomous driving system to relinquish driving control to the human driver.”

In addition to the background information obtained for this patent, NewsRx journalists also obtained the inventors’ summary information for this patent: “In light of the foregoing background, the following presents a simplified summary of the present disclosure in order to provide a basic understanding of some aspects of the invention. This summary is not an extensive overview of the invention. It is not intended to identify key or critical elements of the invention or to delineate the scope of the invention. The following summary merely presents some concepts of the invention in a simplified form as a prelude to the more detailed description provided below.

“In some aspects, a decision system may include at least one processor and a memory unit storing computer-executable instructions. The computer-executable instructions may, in some embodiments, include a machine learning algorithm, such as a neural network. The decision system may be configured to, in operation, make a determination regarding whether a human driver can safely take control of the operation of a vehicle, or whether an autonomous driving system should continue to control the operation of the vehicle. The decision system may be configured to cause the autonomous driving system to relinquish control of the vehicle to the human driver.

“In some aspects, the decision system may characterize the human driver and/or the autonomous driving system and use these characterizations in determining whether the human driver or the autonomous driving system should be given driving control of the vehicle.

“Of course, the methods and systems of the above-referenced embodiments may also include other additional elements, steps, computer-executable instructions, or computer-readable data structures. In this regard, other embodiments are disclosed and claimed herein as well. The details of these and other embodiments of the present invention are set forth in the accompanying drawings and the description below. Other features and advantages of the invention will be apparent from the description, drawings, and claims.”

The claims supplied by the inventors are:

“1. A method comprising: determining, by a computing device, that an autonomous driving system is driving a vehicle with a human driver therein; determining that an unexpected event associated with a conflict has occurred in a driving scenario; retrieving a model that captures attributes personalized to the human driver, wherein the model is based on a plurality of weighted feature functions, wherein each weighted feature function outputs a numerical value associated with a deviation from a policy computed for the human driver associated with at least one of speed, heading, or distance to a lane marker; based on setting a current vehicle state as an input of the model, generating a simulated driving action, including a predicted trajectory; determining that the human driver can safely handle the unexpected event based on the predicted trajectory passing a safety threshold; and causing driving control of the vehicle to be passed to the human driver.

“2. The method of claim 1, wherein an optimal policy is based on a probability distribution of driving actions given a specific vehicle state.

“3. The method of claim 2, wherein each weighted feature function comprises a score associated with a driving action taken by the human driver.

“4. The method of claim 1, wherein the model of the human driver calculates a reaction time or an indication of negotiation skills.

“5. The method of claim 1, wherein the determining that the human driver can safely handle the unexpected event comprises determining, by a neural network, that the human driver can safely handle the unexpected event.

“6. The method of claim 5, further comprising: training the neural network based on labels generated by a rules-based guard sub-system.

“7. The method of claim 6, further comprising: training the neural network by correcting an output of the rules-based guard sub-system, wherein the correcting is based on an observation of results of a previous decision made by the rules-based guard sub-system.

“8. The method of claim 1, wherein the unexpected event comprises: an incursion into a lane occupied by the vehicle, a stationary object in a path of the vehicle, or a loss of critical sensors.

“9. The method of claim 1, wherein the determining that the human driver can safely handle the unexpected event comprises: determining that a reaction time of the human driver is less than a predicted time to an unsafe vehicle position.

“10. A computing platform, comprising: at least one processor; a communication interface; and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: determine, by a computing device, that an autonomous driving system is driving a vehicle with a human driver therein; determine that an unexpected event associated with a conflict has occurred in a driving scenario; retrieve a model that captures attributes personalized to the human driver, wherein the model is based on a plurality of weighted feature functions, wherein each weighted feature function outputs a numerical value associated with a deviation from a policy computed for the human driver associated with at least one of speed, heading, or distance to a lane marker; based on setting a current vehicle state as an input of the model, generate a simulated driving action, including a predicted trajectory; determine that the human driver can safely handle the unexpected event based on the predicted trajectory passing a safety threshold; and cause driving control of the vehicle to be passed to the human driver.

“11. The computing platform of claim 10, wherein an optimal policy is based on a probability distribution of driving actions given a specific vehicle state.

“12. The computing platform of claim 10, wherein the model of the human driver calculates a reaction time or an indication of negotiation skills.

“13. The computing platform of claim 10, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to determine that the human driver can safely handle the unexpected event by causing the computing platform to: determine, by a neural network, that the human driver can safely handle the unexpected event.

“14. The computing platform of claim 13, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: train the neural network based on labels generated by a rules-based guard sub-system.

“15. The computing platform of claim 14, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: train the neural network by correcting an output of the rules-based guard sub-system, wherein the correcting is based on an observation of results of a previous decision made by the rules-based guard sub-system.

“16. The computing platform of claim 10, wherein the unexpected event comprises: an incursion into a lane occupied by the vehicle, a stationary object in a path of the vehicle, or a loss of critical sensors.

“17. The computing platform of claim 10, wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, cause the computing platform to determine that the human driver can safely handle the unexpected event by causing the computing platform to: determine that a reaction time of the human driver is less than a predicted time to an unsafe vehicle position.

“18. One or more non-transitory computer-readable media storing instructions that, when executed by a computing platform comprising at least one processor, a communication interface, and memory, cause the computing platform to: determine, by a computing device, that an autonomous driving system is driving a vehicle with a human driver therein; determine that an unexpected event associated with a conflict has occurred in a driving scenario; retrieve a model that captures attributes personalized to the human driver, wherein the model is based on a plurality of weighted feature functions, wherein each weighted feature function outputs a numerical value associated with a deviation from a policy computed for the human driver associated with at least one of speed, heading, or distance to a lane marker; based on setting a current vehicle state as an input of the model, generate a simulated driving action, including a predicted trajectory; determine that the human driver can safely handle the unexpected event based on the predicted trajectory passing a safety threshold; and cause driving control of the vehicle to be passed to the human driver.”

URL and more information on this patent, see: Aragon, Juan Carlos. Safe hand-off between human driver and autonomous driving system. U.S. Patent Number 11940790, filed December 11, 2019, and published online on March 26, 2024. Patent URL (for desktop use only): https://ppubs.uspto.gov/pubwebapp/external.html?q=(11940790)&db=USPAT&type=ids

(Our reports deliver fact-based news of research and discoveries from around the world.)