System and method for automatically detecting key behaviors by vehicles

Abstract

Aspects of the disclosure relate generally to detecting discrete actions by traveling vehicles. The features described improve the safety, use, driver experience, and performance of autonomously controlled vehicles by performing a behavior analysis on mobile objects in the vicinity of an autonomous vehicle. Specifically, an autonomous vehicle is capable of detecting and tracking nearby vehicles and is able to determine when these nearby vehicles have performed actions of interest by comparing their tracked movements with map data.

The invention claimed is: 
     
       1. A method comprising:
 controlling, by one or more computing devices, an autonomous vehicle in accordance with a first control strategy;   receiving, by the one or more computing devices, sensor data indicating a detection of a first object;   classifying, by the one or more computing devices, the first object based on the sensor data;   accessing, by the one or more computing devices, behavior data based on a classification of the first object, wherein the behavior data identifies potential actions of the first object that are to result in a change in control strategy, and wherein at least one of the potential actions identified in the behavior data is the action of changing from traveling on a first road element to travelling on a second road element;   determining, by the one or more computing devices, that the first object has performed an action identified in the behavior data; and   based on the determination, altering the control strategy of the autonomous vehicle by the one or more computing devices.   
     
     
       2. The method of  claim 1 , wherein the first road element is a first lane of traffic on a roadway, and wherein the second road element is a second lane of traffic on the roadway. 
     
     
       3. The method of  claim 1 , wherein altering the control strategy of the autonomous vehicle comprises positioning the autonomous vehicle relative to the first object in a predefined manner. 
     
     
       4. The method of  claim 1 , wherein altering the control strategy of the autonomous vehicle comprises having the autonomous vehicle change from travelling in a first lance of traffic to travelling in a second lane of traffic. 
     
     
       5. The method of  claim 1 , wherein altering the control strategy of the autonomous vehicle comprises altering at least one of a position, heading, speed, and acceleration of the autonomous vehicle. 
     
     
       6. The method of  claim 1 , wherein the first object is classified as a vehicle. 
     
     
       7. The method of  claim 6 , wherein determining that the first object has performed an action identified in the behavior data further comprises determining that the first object has changed from travelling on the first road element to travelling on the second road element. 
     
     
       8. A method comprising:
 controlling, by one or more computing devices, an autonomous vehicle;   receiving, by the one or more computing devices, sensor data indicating a position of a first object external to the autonomous vehicle;   classifying, by the one or more computing devices, the first object based on the sensor data;   accessing, by the one or more computing devices, map data having a plurality of road elements;   comparing the sensor data with the map data;   identifying, by the one or more computing devices, that the first object is travelling on a first road element from the plurality of road elements;   determining, by the one or more computing devices, that based on the comparison of the sensor data with the map data, the first object has travelled from the first road element to a second road element; and   altering, by the one or more computing devices, at least one of a position, heading, speed, and acceleration of the autonomous vehicle based on the determination that the first object has travelled from the first road element to the second road element.   
     
     
       9. The method of  claim 8 , wherein the first road element is a first lane of traffic on a roadway, and wherein the second road element is a second lane of traffic on the roadway. 
     
     
       10. The method of  claim 8 , wherein altering at least one of the position, heading, speed, and acceleration of the autonomous vehicle comprises positioning the autonomous vehicle relative to the first object in a predefined manner. 
     
     
       11. The method of  claim 8 , further comprising:
 receiving, by the one or more computing devices, a request to navigate between a first location and a second location; and   autonomously navigating, by the one or more computing devices, the autonomous vehicle along a path between the first location and a second location; and   wherein altering at least one of a position,   heading, and speed of the autonomous vehicle, occurs while the autonomous vehicle is travelling along the path.   
     
     
       12. The method of  claim 8 , further comprising:
 determining a relative position of the autonomous vehicle with each of the one or more nearby vehicles, and   wherein associating each of the one or more nearby vehicles with a road graph element is based on the relative position.   
     
     
       13. A system for controlling an autonomous vehicle, the system comprising:
 one or more sensors for detecting a one or more vehicles in an autonomous vehicle's surroundings; and   one or more processors configured to:   control an autonomous vehicle in accordance with a first control strategy;   receive sensor data indicating a detection of a first object;   classify the first object based on the sensor data;   access behavior data based on a classification of the first object, wherein the behavior data identifies potential actions of the first object that are to result in a change in control strategy, and wherein at least one of the potential actions of the first object is the first object changing from traveling on a first road element to travelling on a second road element;   determine that the first object has performed an action identified in the behavior data; and   alter the control strategy of the autonomous vehicle based on the determination.   
     
     
       14. The system of  claim 13 , wherein the first road element is a first lane of traffic on a roadway, and wherein the second road element is a second lane of traffic on the roadway. 
     
     
       15. The system of  claim 13 , wherein altering the control strategy of the autonomous vehicle comprises positioning the autonomous vehicle relative to the first object in a predefined manner. 
     
     
       16. The system of  claim 13 , wherein altering the control strategy of the autonomous vehicle comprises having the autonomous vehicle change from travelling in a first lance of traffic to travelling in a second lane of traffic. 
     
     
       17. The system of  claim 13 , wherein altering the control strategy of the autonomous vehicle comprises altering at least one of a position, heading, speed, and acceleration of the autonomous vehicle. 
     
     
       18. The system of  claim 13 , wherein the sensor data includes information relating to at least one of a position, heading, speed, and acceleration of the first object. 
     
     
       19. The system of  claim 18 , wherein determining that the first object has performed an action identified in the behavior data further comprises determining that the first object has changed from travelling on the first road element to travelling on the second road element. 
     
     
       20. The system of  claim 13 , wherein the one or more processors are further configured to:
 receive a request for navigation between a first location and a second location; and   autonomously navigate the autonomous vehicle along a path between the first location and a second location; and   wherein altering the control strategy occurs while the autonomous vehicle travels along the path.   
     
     
       21. A method comprising:
 receiving, by one or more sensors of a vehicle configured to operate in a fully autonomous mode, sensor data of an external environment of the vehicle;   filtering, by one or more computing devices of the vehicle, the received sensor data to identify one or more actions of interest of an object in the external environment, the filtering including filtering the received sensor data to only include instances where the object has performed an action of interest;   determining, by the one or more computing devices, whether the vehicle cannot currently operate in the fully autonomous mode based on the identified one or more actions of interest; and   upon determining by the one or more computing devices that the vehicle cannot currently operate in the fully autonomous mode, the one or more computing devices altering a control strategy;   wherein altering the control strategy includes either (i) delaying changing to the fully autonomous mode, or (ii) ceding control of the vehicle to a remote operator.    
     
     
       22. The method of claim 21, wherein the object is another vehicle.  
     
     
       23. The method of claim 21, wherein delaying changing to the fully autonomous mode includes the one or more computing devices scanning the external environment to determine whether there are any obstacles affecting an ability of the vehicle to avoid a collision.  
     
     
       24. The method of claim 21, wherein delaying changing to the fully autonomous mode includes the one or more computing devices requiring a driver of the vehicle to control steering or accelerating before entering into the fully autonomous mode.  
     
     
       25. The method of claim 21, wherein ceding control of the vehicle to the remote operator includes sending the received sensor data to a remote party associated with the remote operator.  
     
     
       26. The method of claim 25, further comprising transmitting data or imagery to a remote computing device in conjunction with ceding control to the remote operator.  
     
     
       27. The method of claim 21, wherein the identified one or more actions of interest include another vehicle changing lanes.  
     
     
       28. The method of claim 21, wherein the identified one or more actions of interest include another vehicle changing its route.