Systems and methods for automatic routing and autonomous driving using driving skill evaluations
Systems and methods for evaluating driving performance for a given driver/user of a motor vehicle, and automatically modifying the operation of one or more vehicle systems based on the evaluated driving performance. Driving performance may be inferred from input from one or more sensors. The sensor input may be fed to a machine learning classifier, which predicts a corresponding “driving skill level,” such as “Beginner,” “Intermediate,” or “Advanced.” Using the predicted driving skill level, the performance of vehicle systems may be modified to improve safety and/or driver experience.
1 . A motor vehicle, comprising: a plurality of sensors; an autonomous driving system; a navigation system; and a driving skill system, wherein the driving skill system is further configured to: receive information from the plurality of sensors while a user is driving the motor vehicle; provide the information from the plurality of sensors to a machine learning classifier to determine a driving skill level for the user; and modify the operation of the autonomous driving system using at least the driving skill level, wherein modifying the operation of the vehicle system comprises: determining that all of a multiple of routes generated by the navigation system from a current location to a destination selected by the user are too difficult for the user according to the driving skill level of the user; determining that a route from the current location to the destination is suitable for the autonomous driving system of the motor vehicle; and engaging the autonomous driving system to autonomously drive the motor vehicle along the route suitable for the autonomous driving system.
2 . The motor vehicle according to claim 1 , wherein the information from the plurality of sensors includes one or more of: gyroscope information, vehicle velocity information, vehicle acceleration information, steering sensor information, throttle sensor information, brake sensor information, gear state information, LIDAR information, RADAR information, or vehicle camera information.
3 . The motor vehicle according to claim 1 , wherein the determining that all of the multiple of routes are too difficult for the user further comprises determining a route difficulty for each of the multiple of routes, and wherein the route difficulty is determined based on one or more of: speed limits along the route, whether merging is required, traffic levels, or a need to navigate round-abouts or U-turns.
4 . The motor vehicle according to claim 1 , wherein the machine learning classifier is configured to output three or more different classifications.
5 . The motor vehicle according to claim 1 , wherein the machine learning classifier runs on a computing system integrated into the motor vehicle.
6 . The motor vehicle according to claim 1 , wherein the machine learning classifier is a neural network.
7 . A method of controlling a motor vehicle, comprising: receiving information from a plurality of sensors while a user is driving the motor vehicle; providing the information from a plurality of sensors of the motor vehicle to a machine learning classifier to determine a driving skill level for the user; and modifying the operation of an autonomous driving system of the motor vehicle, using at least the driving skill level, by: determining that all of a multiple of routes generated by a navigation system from a current location to a destination selected by the user are too difficult for the user according to the driving skill level of the user; determining that a route from the current location to the destination is suitable for an autonomous driving system of the motor vehicle; and engaging the autonomous driving system to autonomously drive the motor vehicle along the route suitable for the autonomous driving system.
8 . The method according to claim 7 , wherein the machine learning classifier runs on a computing system integrated into the motor vehicle.
9 . The method according to claim 7 , wherein the machine learning classifier is a neural network.
10 . The method according to claim 7 , wherein the information from the plurality of sensors includes one or more of: gyroscope information, vehicle velocity information, vehicle acceleration information, steering sensor information, throttle sensor information, brake sensor information, gear state information, LIDAR information, RADAR information, or vehicle camera information.
11 . The method according to claim 7 , wherein the machine learning classifier is configured to output three or more different classifications.
12 . The method according to claim 11 , wherein the three or more classifications include a first classification corresponding to a beginner level of driving, a second classification corresponding to an intermediate level of driving, and a third classification corresponding to an advanced level of driving.
13 . A system for controlling a motor vehicle, comprising: an autonomous driving system; and a driving skill system, wherein the driving skill system is further configured to: receive information from a plurality of sensors while a user is driving the motor vehicle; provide the information from the plurality of sensors to a machine learning classifier to determine a driving skill level for the user; and modify the operation of the autonomous driving system using at least the driving skill level, wherein modifying the operation of the vehicle system comprises: determining that all of a multiple of routes generated by a navigation system from a current location to a destination selected by the user are too difficult for the user according to the driving skill level of the user; determining that a route from the current location to the destination is suitable for the autonomous driving system of the motor vehicle; and engaging the autonomous driving system to autonomously drive the motor vehicle along the route suitable for the autonomous driving system.
14 . The system according to claim 13 ,
wherein determining that the route is suitable for the autonomous driving system comprises determining a driving skill level for the autonomous driving system and comparing the driving skill level for the autonomous driving system to a route difficulty for the route.
15 . The system according to claim 13 , wherein the driving skill level for the autonomous driving system is provided by a manufacturer.
16 . The system according to claim 13 , wherein the driving skill level for the autonomous driving system is determined by evaluating driving performance of the autonomous driving system.
17 . The system according to claim 13 , wherein the machine learning classifier is a neural network.
18 . The system according to claim 17 , wherein the neural network includes at least one of a recurrent neural network, a convolution neural network, or a transformer.
19 . The system according to claim 13 , wherein the machine learning classifier is configured to output three or more different classifications.
20 . The system according to claim 13 , wherein the information from the plurality of sensors includes one or more of: gyroscope information, vehicle velocity information, vehicle acceleration information, steering sensor information, throttle sensor information, brake sensor information, gear state information, LIDAR information, RADAR information, or vehicle camera information.