Psychomotor vigilance testing for persons tasked with monitoring autonomous vehicles

Abstract

Assessing a likelihood of a person experiencing a fatigue event when the person tasked with monitoring a vehicle operating in an autonomous driving mode may include receiving a set of response times for a psychomotor vigilance test administered to the person. The test may include a plurality of trials which involve a person lifting a finger from a user input device. Whether the person passed or failed each trial of the set of trials may be determined. A model trained using data from prior psychomotor vigilance tests administered to the person may be identified for the person. Results of the determinations of whether the person passed or failed each trial of the set of trials may be input into the model in order to determine a value representative of a likelihood of a fatigue event. An intervention response may be initiated based on the value.

1 . A method, comprising:
 accessing, by one or more server computing devices, results of a plurality of first sets of psychomotor vigilance tests (PVTs) administered to a particular person at different points in time, the results including respective response times for the particular person, the particular person being tasked with monitoring a vehicle operating in an autonomous driving mode;   determining, by the one or more server computing devices based on the results, respective scores for each of the plurality of first sets of PVTs;   accessing, by the one or more server computing devices, information from a remote monitoring system identifying respective estimated amounts of fatigue of the particular person at different times;   determining, by the one or more server computing devices, whether the information indicates that the particular person experienced one or more first fatigue events; and   training, by the one or more server computing devices based on the scores and determining whether the information indicates that the particular person experienced the one or more first fatigue events, a model individualized to the particular person such that the model outputs a value indicative of a likelihood of the particular person experiencing a second fatigue event in response to inputting data from a second set of PVTs into the model.   
     
     
         2 . The method of  claim 1 , wherein accessing the results includes retrieving, by the one or more server computing devices, the results from a remote monitoring system. 
     
     
         3 . The method of  claim 1 , wherein accessing the results includes accessing, by the one or more server computing devices, self-reported data for the particular person. 
     
     
         4 . The method of  claim 1 , wherein accessing the results includes accessing, by the one or more server computing devices, a plurality of scores associated with the plurality of first sets of PVTs. 
     
     
         5 . The method of  claim 4 , wherein the plurality of scores represents a passing or failing rate for the particular person. 
     
     
         6 . The method of  claim 1 , wherein accessing the results includes include accessing, by the one or more server computing devices, date and time information associated with the plurality of first sets of PVTs. 
     
     
         7 . The method of  claim 1 , wherein training the model includes providing, by the one or more server computing devices, one or more parameter values for the model, the one or more parameter values being used by the model to predict one or more future fatigue events for the particular person. 
     
     
         8 . The method of  claim 1 , wherein accessing the information includes accessing, by the one or more server computing devices, data identifying where the particular person is with respect to his or her circadian rhythm. 
     
     
         9 . The method of  claim 1 , wherein accessing the information includes accessing, by the one or more server computing devices, data identifying a relative point in time for a shift for monitoring the vehicle of the particular person. 
     
     
         10 . The method of  claim 1 , wherein accessing the information includes accessing, by the one or more server computing devices, data identifying an amount of time since a last break of the particular person. 
     
     
         11 . The method of  claim 1 , wherein accessing the information includes accessing, by the one or more server computing devices, data corresponding to an amount of time that the particular person has spent monitoring the vehicle uninterrupted. 
     
     
         12 . The method of  claim 1 , wherein the value is on a scale of 0 to 1 representing the likelihood of the particular person experiencing the second fatigue event. 
     
     
         13 . The method of  claim 1 , further comprising associating, by the one or more server computing devices, the model with the particular person using an identifying code. 
     
     
         14 . The method of  claim 1 , further comprising, subsequent to training the model, by the one or more server computing devices, storing the model in memory accessible by the one or more server computing devices. 
     
     
         15 . A system, comprising:
 one or more processors configured to:   access results of a plurality of first sets of psychomotor vigilance tests (PVTs) administered to a particular person at different points in time, the results including respective response times for the particular person, the particular person being tasked with monitoring a vehicle operating in an autonomous driving mode;   determine, based on the results, respective scores for each of the plurality of first sets of PVTs;   access information from a remote monitoring system identifying respective estimated amounts of fatigue of the particular person at different times;   determine whether the information indicates that the particular person experienced one or more first fatigue events; and   train, based on the scores and the determination whether the information indicates that the particular person experienced the one or more first fatigue events, a model individualized to the particular person such that the model outputs a value indicative of a likelihood of the particular person experiencing a second fatigue event in response to inputting data from a second set of PVTs into the model.   
     
     
         16 . The system of  claim 15 , wherein the one or more processors are further configured to retrieve the results from a remote monitoring system. 
     
     
         17 . The system of  claim 15 , wherein the one or more processors are further configured to access self-reported data for the particular person, the results including the self-reported data. 
     
     
         18 . The system of  claim 15 , wherein the one or more processors are further configured to, in association with training of the model, provide one or more parameter values for the model, the one or more parameter values being used by the model to predict one or more future fatigue events for the particular person. 
     
     
         19 . The system of  claim 15 , wherein the one or more processors are further configured to associate the model with the particular person using an identifying code. 
     
     
         20 . The system of  claim 15 , wherein the one or more processors are further configured to, subsequent to training of the model, store the model in memory accessible by the one or more processors.