Semantic mapping of environments for autonomous devices
Methods, systems, and apparatus for receiving a reference to an object located in an environment of a robot, accessing mapping data that indicates, for each of a plurality of object instances, respective probabilities of the object instance being located at one or more locations in the environment, wherein the respective probabilities are based at least on an amount of time that has passed since a prior observation of the object instance was made, identifying one or more particular object instances that correspond to the referenced object, determining, based at least on the mapping data, the respective probabilities of the one or more particular object instances being located at the one or more locations in the environment, selecting, based at least on the respective probabilities, a particular location in the environment where the referenced object is most likely located, and directing the robot to navigate to the particular location.
1. A method, comprising:
receiving a reference to a particular object in an environment of a mobile robot;
accessing data that identifies a known object and probability information for the known object, wherein the probability information indicates probabilities of the known object being located at various locations in the environment;
determining, by a processor, a match between the particular object and the known object;
in response to determining a match between the particular object and the known object, using the probability information for the known object to predict a first location of the particular object;
determining, by the processor, that the particular object is not at the first location, and in response, using the probability information for the known object to predict a second location of the particular object; and
directing the robot to navigate to the second location.
2. The method of claim 1 , wherein the data that identifies the known object and the probability information for the known object comprises a map of the environment.
3. The method of claim 1 , wherein the probability information for the known object is based on prior observations of locations of the known object in the environment.
4. The method of claim 3 , comprising adjusting the probability information for the known object based on ages of the prior observations.
5. The method of claim 1 , wherein using the probability information for the known object to predict the first location of the particular object comprises selecting as the first location a location in the environment where the known object is assigned the highest probability of being located.
6. The method of claim 5 , wherein using the probability information for the known object to predict the second location of the particular object comprises selecting as the second location a location in the environment where the known object is assigned a probability lower than the highest probability.
7. The method of claim 1 , wherein determining that the particular object is not at the first location comprises navigating the robot to the first location, and using the robot to acquire an observation of the first location that indicates the robot is not at the first location.
8. The method of claim 1 , comprising using the robot to interact with the particular object upon the robot's arrival at the second location.
9. The method of claim 1 , comprising updating the probability information for the known object based on determining that the particular object is not at the first location.
10. The method of claim 1 , wherein the probability information for the known object is based on spatial relationships between the known object and other known objects in the environment.
11. A system comprising:
one or more processors configured to execute computer program instructions; and
one or more computer-storage media encoded with computer programs that, when
executed by the one or more processors, cause the system to perform operations comprising:
receiving a reference to a particular object in an environment of a mobile robot;
accessing data that identifies a known object and probability information for the known object, wherein the probability information indicates probabilities of the known object being located at various locations in the environment;
determining a match between the particular object and the known object;
in response to determining a match between the particular object and the known object, using the probability information for the known object to predict a first location of the particular object;
determining that the particular object is not at the first location, and in response, using the probability information for the known object to predict a second location of the particular object; and
directing the robot to navigate to the second location.
12. The system of claim 11 , wherein the data that identifies the known object and the probability information for the known object comprises a map of the environment.
13. The system of claim 11 , wherein the probability information for the known object is based on prior observations of locations of the known object in the environment.
14. The system of claim 13 , wherein the operations comprise adjusting the probability information for the known object based on ages of the prior observations.
15. The system of claim 11 , wherein using the probability information for the known object to predict the first location of the particular object comprises selecting as the first location a location in the environment where the known object is assigned the highest probability of being located.
16. The system of claim 15 , wherein using the probability information for the known object to predict the second location of the particular object comprises selecting as the second location a location in the environment where the known object is assigned a probability lower than the highest probability.
17. The system of claim 11 , wherein determining that the particular object is not at the first location comprises navigating the robot to the first location, and using the robot to acquire an observation of the first location that indicates the robot is not at the first location.
18. The system of claim 11 , wherein the operations comprise using the robot to interact with the particular object upon the robot's arrival at the second location.
19. The system of claim 11 , wherein the operations comprise updating the probability information for the known object based on determining that the particular object is not at the first location.
20. One or more computer-readable devices storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
receiving a reference to a particular object in an environment of a mobile robot;
accessing data that identifies a known object and probability information for the known object, wherein the probability information indicates probabilities of the known object being located at various locations in the environment;
determining a match between the particular object and the known object;
in response to determining a match between the particular object and the known object, using the probability information for the known object to predict a first location of the particular object;
determining that the particular object is not at the first location, and in response, using the probability information for the known object to predict a second location of the particular object; and
directing the robot to navigate to the second location.