Intention recognition

Abstract

Some embodiments provide an autonomous navigation system which autonomously navigates a vehicle through an environment based on predicted trajectories of one or more separate dynamic elements through the environment. The system identifies contextual cues associated with a monitored dynamic element, based on features of the dynamic element and actions of the dynamic element relative to various elements of the environment, including motions relative to other dynamic elements. A monitored dynamic element can be associated with a particular intention, which specifies a prediction of dynamic element movement through the environment, based on a correlation between identified contextual cues associated with the monitored dynamic element and a set of contextual cues which are associated with the particular intention. A predicted trajectory of the dynamic element is generated based on an associated intention. A targeted signal can be directed to a target dynamic element based on a predicted trajectory of the dynamic element.

What is claimed is: 
     
         1 . An apparatus, comprising:
 an autonomous navigation system configured to be installed in a vehicle and autonomously navigate the vehicle through an environment in which the vehicle is located, wherein the autonomous navigation system is configured to:
 identify a set of contextual cues associated with a dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the dynamic element, based on monitoring at least a portion of the environment; 
 associate the dynamic element with a particular set of predicted motions, based on a determination of a correlation between the identified set of contextual cues and a predetermined set of contextual cues which are associated with the particular set of predicted motions; 
 generate a predicted trajectory of the dynamic element through the environment based on the particular set of predicted motions associated with the dynamic element; and 
 generate a set of control commands which, when executed by one or more control elements installed in the vehicle, cause the vehicle to be navigated along a driving route that accounts for the predicted trajectory of the dynamic element. 
   
     
     
         2 . The apparatus of  claim 1 , wherein the autonomous navigation system is configured to:
 prior to identifying the set of contextual cues associated with the dynamic element located in the environment:
 identify a set of contextual cues associated with another dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the other dynamic element, based on monitoring at least a portion of the environment; 
 monitor a particular set of motions executed by the other dynamic element through the environment, subsequent to identifying the set of contextual cues associated with the other dynamic element; and 
 based on the monitoring, generate an association of the set of contextual cues, as the set of predetermined contextual cues, with the particular set of motions executed by the other dynamic element, as the particular set of predicted motions; 
 wherein the association specifies that dynamic elements associated with the set of predetermined contextual cues are predicted to execute the particular set of predicted motions. 
   
     
     
         3 . The apparatus of  claim 2 , wherein the autonomous navigation system is configured to:
 monitor a particular set of actual motions executed by the dynamic element through the environment, subsequent to associating the dynamic element with the particular set of predicted motions; and   based on a determination that the particular set of actual motions executed by the dynamic element are distinct from the particular set of predicted motions, associate the predetermined set of contextual cues with the particular set of actual motions and disassociating the predetermined set of contextual cues from the particular set of predicted motions.   
     
     
         4 . The apparatus of  claim 1 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to one or more particular static elements detected in the environment.   
     
     
         5 . The apparatus of  claim 1 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to the one or more other dynamic elements detected in the environment.   
     
     
         6 . The apparatus of  claim 1 , comprising:
 one or more sets of signal generators which are configured to generate and direct one or more targeted signals to one or more elements located in the environment; and   an autonomous navigation system configured to:
 command at least one of the one or more sets of signal generators to generate a particular targeted signal, comprising a particular instance of content selected based on at least one sensor data representation of the dynamic element and a particular signal beam angle which intersects a limited portion, of the environment, in which the particular element is located, such that the particular instance of content is communicated to the dynamic element in the environment to the exclusion of a remainder portion of the environment. 
   
     
     
         7 . The apparatus of  claim 6 , wherein the one or more sets of signal generators comprise one or more sets of speaker devices which are configured to generate one or more targeted signals to one or more elements located in the environment based on at least one of:
 beamforming; or   ultrasonic modulation.   
     
     
         8 . A method, comprising:
 performing, by one or more computer systems installed in a vehicle located in a particular environment:
 identifying a set of contextual cues associated with a dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the dynamic element, based on monitoring at least a portion of the environment; 
 associating the dynamic element with a particular set of predicted motions, based on a determination of a correlation between the identified set of contextual cues and a predetermined set of contextual cues which are associated with the particular set of predicted motions; 
 generating a predicted trajectory of the dynamic element through the environment based on the particular set of predicted motions associated with the dynamic element; and 
 generating a set of control commands which, when executed by one or more control elements installed in the vehicle, cause the vehicle to be navigated along a driving route that accounts for the predicted trajectory of the dynamic element. 
   
     
     
         9 . The method of  claim 8 , comprising:
 prior to identifying the set of contextual cues associated with the dynamic element located in the environment:
 identifying a set of contextual cues associated with another dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the other dynamic element, based on monitoring at least a portion of the environment; 
 monitoring a particular set of motions executed by the other dynamic element through the environment, subsequent to identifying the set of contextual cues associated with the other dynamic element; and 
 based on the monitoring, generating an association of the set of contextual cues, as the set of predetermined contextual cues, with the particular set of motions executed by the other dynamic element, as the particular set of predicted motions; 
 wherein the association specifies that dynamic elements associated with the set of predetermined contextual cues are predicted to execute the particular set of predicted motions. 
   
     
     
         10 . The method of  claim 9 , comprising:
 monitoring a particular set of actual motions executed by the dynamic element through the environment, subsequent to associating the dynamic element with the particular set of predicted motions; and   based on a determination that the particular set of actual motions executed by the dynamic element are distinct from the particular set of predicted motions, associating the predetermined set of contextual cues with the particular set of actual motions and disassociating the predetermined set of contextual cues from the particular set of predicted motions.   
     
     
         11 . The method of  claim 8 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to one or more particular static elements detected in the environment.   
     
     
         12 . The method of  claim 8 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to the one or more other dynamic elements detected in the environment.   
     
     
         13 . The method of  claim 8 , comprising:
 commanding at least one set of one or more signal generators installed in the vehicle to generate a particular targeted signal, comprising a particular instance of content selected based on at least one sensor data representation of the dynamic element and a particular signal beam angle which intersects a limited portion, of the environment, in which the particular element is located, such that the particular instance of content is communicated to the dynamic element in the environment to the exclusion of a remainder portion of the environment.   
     
     
         14 . The method of  claim 13 , wherein the one or more sets of signal generators comprise one or more sets of speaker devices which are configured to generate one or more targeted signals to one or more elements located in the environment based on at least one of:
 beamforming; or   ultrasonic modulation.   
     
     
         15 . A non-transitory, computer-readable medium storing a program of instructions which, when executed by at least one computer system, causes the at least one computer system to:
 identify a set of contextual cues associated with a dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the dynamic element, based on monitoring at least a portion of the environment;   associate the dynamic element with a particular set of predicted motions, based on a determination of a correlation between the identified set of contextual cues and a predetermined set of contextual cues which are associated with the particular set of predicted motions;   generate a predicted trajectory of the dynamic element through the environment based on the particular set of predicted motions associated with the dynamic element; and   generate a set of control commands which, when executed by one or more control elements installed in the vehicle, cause the vehicle to be navigated along a driving route that accounts for the predicted trajectory of the dynamic element.   
     
     
         16 . The non-transitory, computer readable medium of  claim 15 , wherein the program of instruction, when executed by the at least one computer system, cause the at least one computer system to:
 prior to identifying the set of contextual cues associated with the dynamic element located in the environment:
 identify a set of contextual cues associated with another dynamic element located in the environment, wherein each contextual cue indicates one or more particular features associated with the other dynamic element, based on monitoring at least a portion of the environment; 
 monitor a particular set of motions executed by the other dynamic element through the environment, subsequent to identifying the set of contextual cues associated with the other dynamic element; and 
 based on the monitoring, generate an association of the set of contextual cues, as the set of predetermined contextual cues, with the particular set of motions executed by the other dynamic element, as the particular set of predicted motions; 
 wherein the association specifies that dynamic elements associated with the set of predetermined contextual cues are predicted to execute the particular set of predicted motions. 
   
     
     
         17 . The non-transitory, computer readable medium of  claim 15 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to one or more particular static elements detected in the environment.   
     
     
         18 . The non-transitory, computer readable medium of  claim 15 , wherein:
 the set of contextual cues comprises a specification of a position and velocity of the dynamic element relative to the one or more other dynamic elements detected in the environment.   
     
     
         19 . The non-transitory, computer readable medium of  claim 15 , wherein the program of instruction, when executed by the at least one computer system, cause the at least one computer system to:
 command at least one set of one or more signal generators installed in the vehicle to generate a particular targeted signal, comprising a particular instance of content selected based on at least one sensor data representation of the dynamic element and a particular signal beam angle which intersects a limited portion, of the environment, in which the particular element is located, such that the particular instance of content is communicated to the dynamic element in the environment to the exclusion of a remainder portion of the environment.   
     
     
         20 . The non-transitory, computer readable medium of  claim 19 , wherein the one or more sets of signal generators comprise one or more sets of speaker devices which are configured to generate one or more targeted signals to one or more elements located in the environment based on at least one of:
 beamforming; or   ultrasonic modulation.