Processing system for evaluating autonomous vehicle control systems through continuous learning

1 . A computing platform, comprising:

at least one processor;

a communication interface communicatively coupled to the at least one processor; and

memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:

determine that an autonomous driving system of an autonomous vehicle has caused evolution of one or more driving patterns;

generate, based on the evolution, one or more learning curves for the autonomous driving system;

determine that the evolution caused the autonomous driving system to fall below a predetermined safety threshold based on the learning curves;

determine, in response to determining that the evolution caused the autonomous driving system to fall below the predetermined safety threshold, a driving pattern degradation output for the autonomous vehicle based on sensor data, the sensor data indicating at least one of:

an attentiveness distribution associated with one or more neighboring vehicles; or

an aggressiveness distribution associated with the one or more neighboring vehicles; and

send the driving pattern degradation output to cause the autonomous driving system of the autonomous vehicle to perform an autonomous action by reducing speed instead of changing lanes in response to the attentiveness distribution or the aggressiveness distribution.

2 . The computing platform of claim 1 , wherein performing the autonomous vehicle action includes reducing speed instead of changing lanes in response to the attentiveness distribution or the aggressiveness distribution.

3 . The computing platform of claim 1 , wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to receive, while the vehicle is operating in an autonomous mode, the sensor data from one or more autonomous vehicle sensors, and the computing platform is an autonomous vehicle evaluation system configured to evaluate the autonomous driving system autonomously operating the vehicle.

4 . The computing platform of claim 1 , wherein the computer-readable instructions, when executed by the at least one processor, further cause the computing platform to:

determine, based on the sensor data, one or more driving patterns; and

group the one or more driving patterns based on a primary context corresponding to the one or more driving patterns, wherein the driving pattern degradation output indicates degradation corresponding to the one or more driving patterns.

5 . The computing platform of claim 1 , wherein the sensor data used for determining the driving pattern degradation output further indicates an identification of the one or more neighboring vehicles as human or as another autonomous driving system.

6 . The computing platform of claim 1 , wherein the sensor data used for determining the driving pattern degradation output further indicates a lane changing pattern of the one or more neighboring vehicles.

7 . The computing platform of claim 1 , wherein the sensor data used for determining the driving pattern degradation output further indicates a front vehicle performing hard braking.

8 . The computing platform of claim 1 , wherein the autonomous vehicle action includes changing an acceleration parameter for a lane change.

9 . The computing platform of claim 1 , wherein the autonomous vehicle action includes changing a lane change acceleration value.

10 . The computing platform of claim 1 , wherein the autonomous vehicle action includes reducing a speed based on the sensor data indicating a presence of one or more aggressive drivers.

11 . A method comprising:

receiving, by a computing device having at least one processor and while an autonomous vehicle is operating in an autonomous mode, sensor data from one or more vehicle sensors;

determining that an autonomous driving system of the autonomous vehicle has caused evolution of one or more driving patterns;

generating, based on the evolution, one or more learning curves for the autonomous driving system;

determining that the evolution caused the autonomous driving system to fall below a predetermined safety threshold based on the learning curves;

determining in response to determining that the evolution caused the autonomous driving system to fall below the predetermined safety threshold, by the at least one processor, a driving pattern degradation output based on the sensor data, the sensor data indicating at least one of:

an attentiveness distribution associated with a plurality of neighboring vehicles; or

an aggressiveness distribution associated with the plurality of neighboring vehicles; and

sending, by the at least one processor, the driving pattern degradation output to cause the autonomous driving system of the vehicle to perform an autonomous action by reducing speed instead of changing lanes in response to the attentiveness distribution or the aggressiveness distribution.

12 . The method of claim 11 , wherein performing the autonomous vehicle action includes reducing speed instead of changing lanes in response to the attentiveness distribution or the aggressiveness distribution.

13 . The method of claim 11 , wherein the autonomous vehicle action includes increasing a lane change acceleration based on the sensor data indicating a presence of a neighboring vehicle behind the vehicle in a target lane.

14 . The method of claim 13 , wherein the sensor data indicates the neighboring vehicle is within six meters of the vehicle.

15 . The method of claim 11 , wherein the autonomous vehicle action includes decreasing a lane change acceleration based on an absence of a neighboring vehicle within a predetermined distance behind the vehicle in a target lane.

16 . The method of claim 11 , wherein the sensor data used for determining the driving pattern degradation output further indicates an identification of a neighboring vehicle as human or as another autonomous driving system.

17 . The method of claim 11 , wherein the sensor data used for determining the driving pattern degradation output further indicates a lane changing pattern of a neighboring vehicle.

18 . The method of claim 11 , wherein the sensor data used for determining the driving pattern degradation output further indicates a front vehicle performing hard braking.

19 . A non-transitory computer readable medium storing computer executable instructions which, when executed by a processor, cause a computing device to perform steps comprising:

determining that an autonomous driving system of an autonomous vehicle has caused evolution of one or more driving patterns;

generating, based on the evolution, one or more learning curves for the autonomous driving system;

determining that the evolution caused the autonomous driving system to fall below a predetermined safety threshold based on the learning curves;

determining, in response to determining that the evolution caused the autonomous driving system to fall below the predetermined safety threshold, a driving pattern degradation output for a vehicle based on sensor data, the sensor data indicating at least one of:

an attentiveness distribution associated with one or more neighboring vehicles; or

an aggressiveness distribution associated with the one or more neighboring vehicles; and

sending the driving pattern degradation output causing the autonomous driving system of the autonomous vehicle to perform an autonomous action by reducing speed instead of changing lanes in response to the attentiveness distribution or the aggressiveness distribution.

20 . The non-transitory computer readable medium of claim 19 , wherein the computer executable instructions, when executed by the processor, further cause the computing device to receive, while the vehicle is operating in an autonomous mode, the sensor data from one or more autonomous vehicle sensors including at least one of a camera, a lidar sensor, or a radar sensor, and the computing device is an autonomous vehicle evaluation system configured to evaluate the autonomous driving system autonomously operating the vehicle.