Manufacturer-based autonomous driving assessment

1 . A system, comprising:

a communication component coupled to an autonomous driving system of an autonomous vehicle and to one or more sensors employed on the autonomous vehicle; and

a processor remote from the autonomous vehicle configured to couple to the communication component, wherein the processor is configured to:

receive a request for an assessment of the autonomous driving system of the autonomous vehicle under a particular one or more conditions;

predict that the autonomous vehicle will experience the particular one or more conditions, by querying one or more databases with a location of the autonomous vehicle, the query returning the particular one or more conditions;

in response to predicting that the autonomous vehicle will experience the particular one or more conditions based on the location of the autonomous vehicle, trigger the assessment for the autonomous driving system by the autonomous vehicle via one or more commands sent by the processor to the autonomous vehicle via the communication component, the one or more commands instructing the autonomous vehicle to collect driving response data from the one or more sensors;

in response to triggering the assessment for the autonomous driving system of the autonomous vehicle, receive driving data associated with the autonomous driving system, wherein the driving data comprises the driving response data measured by the one or more sensors during the assessment, and wherein the driving response data comprises at least one of:

roll, pitch, and yaw values of the autonomous vehicle;

control commands of the autonomous driving system; or

driver intervention data pertaining to the autonomous driving system;

analyze the driving data comprising the driving response data using machine learning, wherein analyzing the driving data comprises a driving performance analysis for identifying attributes of the autonomous driving system based on the driving data, and wherein the attributes comprise at least one of:

a subset of the roll, pitch, and yaw values of the autonomous vehicle based on the control commands of the autonomous driving system;

frenetic control of the autonomous vehicle based on the control commands of the autonomous driving system; or

an amount of manual override intervention by a driver based on the driver intervention data;

determine characteristics of the autonomous driving system using the machine learning based on the identified attributes, wherein the characteristics of the autonomous driving system comprise a reliability of the autonomous driving system under the one or more conditions, and wherein the reliability of the autonomous driving system under the one or more conditions is determined by at least one of:

comparing the subset of the roll, pitch, and yaw values of the autonomous vehicle to acceptability thresholds; or

evaluating an impact of the amount of manual override intervention by the driver based on driver characteristics of the driver;

generate a risk profile, wherein the risk profile comprises the identified attributes and the determined characteristics;

store the generated risk profile in a database; and

provide an electronic response to the request for the assessment of the autonomous driving system of the autonomous vehicle under the particular one or more conditions, the electronic response comprising an indication of the generated risk profile.

2 . The system of claim 1 , wherein the processor comprises one or more artificial intelligence processors.

3 . The system of claim 1 , wherein the assessment comprises one or more autonomous driving tests using the autonomous driving system, wherein each of the one or more autonomous driving tests is monitored by the one or more sensors.

4 . The system of claim 3 , wherein the processor is configured to initiate the one or more autonomous driving tests during the assessment.

5 . The system of claim 1 , wherein the one or more sensors comprise one or more cameras, Light Detection and Ranging (LiDAR) sensors, acoustic sensors, speed or velocity sensors, accelerometer sensors, global positioning system (GPS) sensors, sonar sensors, orientation sensors, temperature sensors, pressure sensors, voltage or current sensors, stress sensors, inertial sensors, or a combination thereof.

6 . The system of claim 1 , wherein the driving data comprises specification data of the autonomous driving system, historical data of the autonomous driving system, traffic data, road condition data, geo location data, weather data, reference data, and driver-related data.

7 . The system of claim 6 , wherein the reference data comprises average driving responses of one or more autonomous driving systems different from the autonomous driving system.

8 . The system of claim 6 , wherein the driver-related data is associated with one or more drivers occupying the autonomous vehicle in one or more autonomous driving tests of the assessment, wherein the driver-related data comprises age, driving experience, locations, physical conditions, driving habits, frequent driving routes, incident reports, driving behaviour, levels of autonomous driving, and safety driving test result.

9 . The system of claim 1 , wherein the processor is configured to generate a recommendation for one or more users based on predicted characteristics associated with an autonomous ride using the machine learning, the risk profile of the autonomous driving system, and risk profiles of one or more different autonomous driving systems.

10 . The system of claim 9 , wherein the recommendation comprises manufacturer information associated with a manufacturer manufacturing the autonomous driving system, where in the manufacturer information comprises make, model and autonomous driving system version of the autonomous vehicle.

11 . The system of claim 9 , wherein the recommendation comprises a recommended insurance premium for the autonomous ride.

12 . A method, comprising:

receiving a request for an assessment for an autonomous driving system of an autonomous vehicle under a particular one or more conditions;

predicting that the autonomous vehicle will experience the particular one or more conditions, by querying one or more databases with a location of the autonomous vehicle, the query returning the particular one or more conditions;

in response to predicting that the autonomous vehicle will experience the particular one or more conditions based on the location of the autonomous vehicle, triggering the assessment for the autonomous driving system by the autonomous vehicle via one or more commands sent by a processor to the autonomous vehicle via a communication component, the communication component is coupled to the processor and the autonomous vehicle, and the processor is remote from the autonomous vehicle;

causing one or more sensors employed on the autonomous vehicle to measure driving response data of the autonomous driving system during the assessment via the one or more commands sent by the processor to the autonomous vehicle via the communication component, the communication component is further coupled to the one or more sensors;

in response to triggering the assessment for the autonomous driving system of the autonomous vehicle, receiving driving data associated with the autonomous driving system via a communication component, wherein the driving data comprises the driving response data measured by the one or more sensors during the assessment, and wherein the driving response data comprises at least one of:

roll, pitch, and yaw values of the autonomous vehicle;

control commands of the autonomous driving system; or

driver intervention data pertaining to the autonomous driving system;

analyzing the driving data comprising the driving response data using machine learning, wherein analyzing the driving data comprises a driving performance analysis for identifying attributes of the autonomous driving system based on the driving data, and wherein the attributes comprise at least one of:

a subset of the roll, pitch, and yaw values of the autonomous vehicle based on the control commands of the autonomous driving system;

frenetic control of the autonomous vehicle based on the control commands of the autonomous driving system; or

an amount of manual override intervention by a driver based on the driver intervention data;

determining characteristics of the autonomous driving system using the machine learning based on the identified attributes, wherein the characteristics of the autonomous driving system comprise a reliability of the autonomous driving system under the one or more conditions, and wherein the reliability of the autonomous driving system under the one or more conditions is determined by at least one of:

comparing the subset of the roll, pitch, and yaw values of the autonomous vehicle to acceptability thresholds;

evaluating an impact of the amount of manual override intervention by the driver based on driver characteristics of the driver;

generating a risk profile, wherein the risk profile comprises the identified attributes and the determined characteristics;

storing the risk profile in a database; and

providing an electronic response to the request for the assessment of the autonomous driving system of the autonomous vehicle under the particular one or more conditions, the electronic response comprising an indication of the generated risk profile.

13 . The method of claim 12 , wherein the one or more sensors are configured to measure the driving data associated with the autonomous driving system by monitoring one or more autonomous driving tests during the assessment, and wherein each of the one or more autonomous driving tests comprises an instance of using a respective autonomous driving mode or algorithm for a specific driving condition.

14 . The method of claim 13 , wherein the determined characteristics further comprise:

driving safety under one or more driving conditions associated with the one or more autonomous driving tests; and

reliability of one or more autonomous driving modes or algorithms associated with the autonomous driving system in the one or more autonomous driving tests.

15 . The method of claim 12 , wherein receiving driving data comprises using the communication component to obtain the driving data from one or more autonomous driving system specification databases, one or more autonomous driving system history databases, one or more road condition databases, one or more weather information databases, one or more driver information databases, one or more user devices, or any combination thereof.

16 . The method of claim 12 , wherein the identified attributes comprise strengths or weaknesses of sensors, actuators, machine learning and artificial intelligence algorithms, and processors to execute the machine learning and artificial intelligence algorithms associated with the autonomous driving system.

17 . A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, are configured to cause one or more processors to perform operations comprising:

receiving a request for an assessment for an autonomous driving system of an autonomous vehicle under a particular one or more conditions;

predicting that the autonomous vehicle will experience the particular one or more conditions, by querying one or more databases with a location of the autonomous vehicle, the query returning the particular one or more conditions;

in response to predicting that the autonomous vehicle will experience the particular one or more conditions based on the location of the autonomous vehicle, triggering the assessment for the autonomous driving system by the autonomous vehicle via one or more commands sent by a processor to the autonomous vehicle via a communication component, the communication component is coupled to the processor and the autonomous vehicle, and the processor is remote from the autonomous vehicle;

initiating one or more autonomous driving tests during the assessment;

causing one or more sensors employed on the autonomous vehicle to measure driving response data of the autonomous driving system associated with the one or more autonomous driving tests via the one or more commands sent by the processor to the autonomous vehicle via the communication component, the communication component is further coupled to the one or more sensors;

in response to triggering the assessment for the autonomous driving system of the autonomous vehicle, receiving driving data associated with the autonomous driving system via a communication component, wherein the driving data comprises the driving response data, and wherein the driving response data comprises at least one of:

roll, pitch, and yaw values of the autonomous vehicle;

control commands of the autonomous driving system; or

driver intervention data pertaining to the autonomous driving system;

analyzing the driving data comprising the driving response data using machine learning, wherein analyzing the driving data comprises a driving performance analysis for identifying attributes of the autonomous driving system based on the driving data, and wherein the attributes comprise at least one of:

a subset of the roll, pitch, and yaw values of the autonomous vehicle based on the control commands of the autonomous driving system;

frenetic control of the autonomous vehicle based on the control commands of the autonomous driving system; or

an amount of manual override intervention by a driver based on the driver intervention data;

determining characteristics of the autonomous driving system using the machine learning based on the identified attributes, wherein the characteristics of the autonomous driving system comprise a reliability of the autonomous driving system under the one or more conditions, and wherein the reliability of the autonomous driving system under the one or more conditions is determined by at least one of:

comparing the subset of the roll, pitch, and yaw values of the autonomous vehicle to acceptability thresholds;

evaluating an impact of the amount of manual override intervention by the driver based on driver characteristics of the driver;

generating a risk profile, wherein the risk profile comprises the identified attributes and the determined characteristics;

storing the risk profile in a database;

providing an electronic response to the request for the assessment of the autonomous driving system of the autonomous vehicle under the particular one or more conditions, the electronic response comprising an indication of the generated risk profile; and

generating a recommendation for one or more users based on the risk profile.

18 . The non-transitory computer-readable medium of claim 17 , wherein the operations comprise:

predicting a set of characteristics associated with an autonomous ride; and

transforming the predicted characteristics into a ride characteristic vector.

19 . The non-transitory computer-readable medium of claim 18 , wherein the operations comprise:

retrieving a plurality of risk profiles associated with a plurality of autonomous driving systems from the database, wherein the plurality of the plurality of autonomous driving systems comprises the autonomous driving system;

transforming the plurality of risk profiles into a plurality of autonomous driving system matrices each comprising one or more sets of characteristic vectors that represent one or more autonomous driving algorithm implemented in one or more autonomous driving mode associated with a respective autonomous driving system of the plurality of the plurality of autonomous driving systems; and

using the machine learning to identify a candidate autonomous driving system for the autonomous ride based on a similarity analysis between the ride characteristic vector and the plurality of autonomous driving system matrices.

20 . The non-transitory computer-readable medium of claim 19 , wherein the similarity analysis comprises a nearest neighbour search to identify a closest characteristic matrix representing the candidate autonomous driving system with respect to the ride characteristic vector representing the autonomous ride, wherein the nearest neighbour search comprises a Euclidean distance calculation, a cosine distance calculation, or a combination thereof.