Dynamically refining markers in an autonomous world model
A system for performing autonomous operations in an operating environment comprising a plurality of objects includes a server computer. The server computer comprises a nontransitory computer readable medium storing detailed representations of the objects in the operating environment. Each detailed representation of an object comprises one or more markers providing knowledge information related to the object. The autonomous device comprises a non-transitory computer readable medium storing a world model that comprises sparse representations of the objects in the operating environment, a processor configured to use the world model to execute tasks requiring use of at least one of the objects, and a networking card for connecting to the server computer. When execution of a new task requires use of a particular object in the operating environment, the autonomous device updates the world model to include the detailed representation of the particular object.
1 . A system for performing autonomous operations in an operating environment comprising a plurality of objects, the system comprising:
a server computer comprising a first non-transitory computer readable medium storing detailed representations of the objects in the operating environment, wherein each detailed representation of an object comprises one or more markers providing knowledge information related to the object and pertaining to how to manipulate the object by an autonomous device;
the autonomous device comprising:
a second non-transitory computer readable medium storing a world model that comprises sparse representations of the objects in the operating environment, wherein the sparse representations of the objects include less information the detailed representations of the objects stored in the first non-transitory computer readable medium,
a processor configured to use the world model to execute tasks requiring use of at least one of the objects, and
a networking card for connecting to the server computer,
wherein, when executing a new task requiring use of a particular object in the operating environment, the autonomous device updates the world model to include the detailed representation of the particular object including the one or more markers.
2 . The system of claim 1 wherein, upon receiving the new task, the autonomous device issues a request to the server computer using the networking card to retrieve the detailed representation of the particular object.
3 . The system of claim 1 , wherein the autonomous device receives the detailed representation of the particular object for the particular object via the networking card via a push message from the server computer.
4 . The system of claim 3 , wherein the push message is transmitted upon detecting that the autonomous device is within a predetermined distance from the particular object associated with the new task.
5 . The system of claim 1 , wherein (a) the sparse representations of the objects in the operating environment each comprise two-dimensional representations of the objects and (b) the detailed representations of the objects each further comprise a three-dimensional representation of the objects.
6 . The system of claim 1 , wherein the autonomous device is a mobile robot.
7 . The system of claim 6 , wherein the detailed representation of the particular object comprises a first marker indicating a location wherein the autonomous device should be parked in order to interact with the particular object.
8 . The system of claim 6 , wherein the autonomous device uses the world model to navigate around objects in the operating environment.
9 . The system of claim 1 , wherein the detailed representation of the particular object comprises a first marker indicating a location on the particular object to be used by the autonomous device when performing the new task.
10 . The system of claim 1 , wherein the detailed representation of the particular object comprises a first marker indicating an object attached to the particular object.
11 . A system for performing autonomous operations in an operating environment comprising a plurality of objects, the system comprising:
an autonomous device comprising:
a first non-transitory computer readable medium storing detailed representations of the objects in the operating environment, wherein each detailed representation of an object comprises one or more markers providing knowledge information related to the object and pertaining to how to manipulate the object by the autonomous device,
a second non-transitory computer readable medium storing a world model that comprises sparse representations of the objects in the operating environment, wherein the sparse representations of the objects include less information than the detailed representations of the objects stored in the first non-transitory computer readable medium,
a processor configured to use the world model to execute tasks requiring use of at least one of the objects, and
wherein, when executing a new task requiring use of a particular object in the operating environment, the autonomous device updates the world model to include the detailed representation of the particular object including the one or more markers.
12 . The system of claim 11 , wherein (a) the sparse representations of the objects in the operating environment each comprise two-dimensional representations of the objects and (b) the detailed representations of the objects each further comprise a three-dimensional representation of the objects.
13 . The system of claim 11 , wherein the autonomous device is a mobile robot.
14 . The system of claim 13 , wherein the detailed representation of the particular object comprises a first marker indicating a location wherein the autonomous device should be parked in order to interact with the particular object.
15 . The system of claim 13 , wherein the autonomous device uses the world model to navigate around objects in the operating environment.
16 . The system of claim 11 , wherein the detailed representation of the particular object comprises a first marker indicating a location on the particular object to be used by the autonomous device when performing the new task.
17 . The system of claim 11 , wherein the detailed representation of the particular object comprises a first marker indicating an object attached to the particular object.
18 . A method for performing autonomous operations in an operating environment comprising a plurality of objects, the method comprising:
storing, in a first memory at a server computer, detailed representations of the objects in the operating environment, wherein the detailed representation of an object comprises one or more markers providing knowledge information related to the object and pertaining to how to manipulate the object by an autonomous device;
storing, in a second memory at the autonomous device, a world model comprising sparse representations of the objects in the operating environment, wherein the sparse representations of the objects include less information than the detailed representations of the objects stored in the first memory;
receiving, by the autonomous device, a new task requiring use of a particular object in the operating environment;
retrieving, by the autonomous device, a detailed representation of the particular object from the server computer in response to receiving the new task;
updating, by the autonomous device, the world model to include the detailed representation of the particular object including the one or more markers; and
utilizing, by the autonomous device, the updated world model to perform the new task.
19 . The method of claim 18 , further comprising:
deleting, by the autonomous device, the detailed representation of the particular object from the second memory after completion of the new task.