Determining acceptable responses for navigating a vehicle that accounts for external conditions of the vehicle

Abstract

Sensor data indicating a substantially 360 degree surrounding of a vehicle is received via a processor included in the vehicle. The sensor data is collected using multiple sensors included in the vehicle. Additionally, an acceptable response time range for a driver of the vehicle to perform an action with the vehicle is obtained, via the processor, based on the sensor data. Additionally, an actual response time for the driver to perform the action is determined, via the processor, based on CAN data collected from a CAN bus included in the vehicle, and not based on the sensor data. Additionally, a determination is made, via the processor, that the actual response time is not within the acceptable response time range. Additionally, a remedial action is caused, via the processor, to be performed in response to the determining that the actual response time is not within the acceptable response time range.

What is claimed is: 
     
         1 . A computer-implemented method comprising:
 providing, by a computing system, training data for a machine learning model based on sensor data of a plurality of vehicles and responses of a plurality of drivers of the plurality of vehicles;   determining, by the computing system, that a response of a driver of a vehicle is abnormal based on sensor data of an environment of the vehicle and the machine learning model; and   causing, by the computing system, a remedial action to be performed based on the response of the driver.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein the training data includes a first training dataset that includes a first set of sensor data of a first vehicle and a first response of a first driver of the first vehicle and a second training dataset that includes a second set of sensor data of the first vehicle and a second response of a second driver of the first vehicle. 
     
     
         3 . The computer-implemented method of  claim 1 , further comprising:
 providing, by the computing system, a training dataset that includes the sensor data of the environment of the vehicle and the response of the driver to the machine learning model; and   updating, by the computing system, the machine learning model based on the training dataset.   
     
     
         4 . The computer-implemented method of  claim 1 , further comprising:
 determining, by the computing system, a speed limit based on a driver profile associated with the driver of the vehicle, wherein the determining that the response of the driver of the vehicle is abnormal is based on the speed limit.   
     
     
         5 . The computer-implemented method of  claim 1 , wherein the remedial action includes a decrease of a speed of the vehicle to less than a predetermined threshold less than a speed limit associated with a location of the vehicle. 
     
     
         6 . The computer-implemented method of  claim 1 , wherein the remedial action includes an alert that indicates the driver has performed an abnormal response and indicates an action for the driver to perform. 
     
     
         7 . The computer-implemented method of  claim 1 , further comprising:
 determining, by the computing system, the driver has performed an abnormal response at least a predetermined number of times; and   updating, by the computing system, a driver profile associated with the driver based on the performance of the abnormal response at least the predetermined number of times.   
     
     
         8 . The computer-implemented method of  claim 1 , further comprising:
 determining, by the computing system, that a second response of a second driver of a second vehicle is normal based on second sensor data and the machine learning model; and   preventing, by the computing system, performance of a remedial action based on the second response of the second driver.   
     
     
         9 . The computer-implemented method of  claim 1 , further comprising:
 determining, by the computing system, the response of the driver of the vehicle based on a control area network bus in the vehicle.   
     
     
         10 . The computer-implemented method of  claim 1 , further comprising:
 performing, by the computing system, an assessment of the driver based on the machine learning model.   
     
     
         11 . A system comprising:
 at least one processor; and   a memory storing instructions that, when executed by the at least one processor, cause the system to perform operations comprising:
 providing training data for a machine learning model based on sensor data of a plurality of vehicles and responses of a plurality of drivers of the plurality of vehicles; 
 determining that a response of a driver of a vehicle is abnormal based on sensor data of an environment of the vehicle and the machine learning model; and 
 causing a remedial action to be performed based on the response of the driver. 
   
     
     
         12 . The system of  claim 11 , wherein the training data includes a first training dataset that includes a first set of sensor data of a first vehicle and a first response of a first driver of the first vehicle and a second training dataset that includes a second set of sensor data of the first vehicle and a second response of a second driver of the first vehicle. 
     
     
         13 . The system of  claim 11 , the operations further comprising:
 providing, by the computing system, a training dataset that includes the sensor data of the environment of the vehicle and the response of the driver to the machine learning model; and   updating the machine learning model based on the training dataset.   
     
     
         14 . The system of  claim 11 , the operations further comprising:
 determining a speed limit based on a driver profile associated with the driver of the vehicle, wherein the determining that the response of the driver of the vehicle is abnormal is based on the speed limit.   
     
     
         15 . The system of  claim 11 , wherein the remedial action includes a decrease of a speed of the vehicle to less than a predetermined threshold less than a speed limit associated with a location of the vehicle. 
     
     
         16 . A non-transitory computer-readable storage medium including instructions that, when executed by at least on processor of a computing system, cause the computing system to perform operations comprising:
 providing training data for a machine learning model based on sensor data of a plurality of vehicles and responses of a plurality of drivers of the plurality of vehicles;   determining that a response of a driver of a vehicle is abnormal based on sensor data of an environment of the vehicle and the machine learning model; and   causing a remedial action to be performed based on the response of the driver.   
     
     
         17 . The non-transitory computer-readable storage medium of  claim 16 , wherein the training data includes a first training dataset that includes a first set of sensor data of a first vehicle and a first response of a first driver of the first vehicle and a second training dataset that includes a second set of sensor data of the first vehicle and a second response of a second driver of the first vehicle. 
     
     
         18 . The non-transitory computer-readable storage medium of  claim 16 , the operations further comprising:
 providing, by the computing system, a training dataset that includes the sensor data of the environment of the vehicle and the response of the driver to the machine learning model; and   updating the machine learning model based on the training dataset.   
     
     
         19 . The non-transitory computer-readable storage medium of  claim 16 , the operations further comprising:
 determining a speed limit based on a driver profile associated with the driver of the vehicle, wherein the determining that the response of the driver of the vehicle is abnormal is based on the speed limit.   
     
     
         20 . The non-transitory computer-readable storage medium of  claim 16 , wherein the remedial action includes a decrease of a speed of the vehicle to less than a predetermined threshold less than a speed limit associated with a location of the vehicle.