DYNAMIC MAP REGENERATION BASED ON SELF-LEARNING COOPERATION-BASED FEEDBACK
Disclosed herein are devices, methods, and systems for map regeneration of an environment. The system includes a map manager configured to maintain a global map of the environment. The system also includes a robot configured to determine a correspondence between a mapped subarea of the environment and a global map of the environment and, based on the correspondence, transmit map data of the mapped subarea to a map manager. The map manager is configured to determine a quality metric of the map data of the mapped subarea with respect to the global map and update the global map with the map data of the mapped subarea based on the quality metric.
1 . A device comprising a processor configured to:
determine a correspondence between a global map of an environment and a mapped subarea of the environment, wherein the mapped subarea is received from a robot within the environment;
determine a quality metric of map data of the mapped subarea with respect to the global map; and
update the global map with the map data of the mapped subarea based on the quality metric.
2 . The device of claim 1 , wherein the processor is configured to determine the quality metric based on a comparison of the map data of the mapped subarea to an second set of map data that at least partially overlaps with the mapped subarea, wherein the second set of map data is received from a second robot in the environment.
3 . The device of claim 2 , wherein the processor is configured to update, based on whether the map data of the mapped subarea satisfies a predefined criterion with respect to the second set of map data, the global map with the map data of the mapped subarea.
4 . The device of claim 3 , wherein the predefined criterion comprises an occupancy metric that characterizes occupied spaces in the mapped subarea and relates to occupancy information of the map data and of the second set of map data that at least partially overlaps with the mapped subarea.
5 . The device of claim 3 , wherein the predefined criterion comprises: an environmental characteristic of the map data with respect to the second set of map data; a floor reflection or a light condition of the mapped subarea; or an extent to which objects in the global map correspond to objects within the map data of the mapped subarea.
6 . The device of claim 1 , wherein the processor configured to update the global map comprises the processor configured to stitch together the map data of the mapped subarea with the global map along an edge of the mapped subarea with an edge of the global map.
7 . The device of claim 1 , wherein the processor configured to determine the correspondence comprises: the processor configured to identify occupied space in the map data of the mapped subarea that differs from the global map; or the processor configured to determine whether the mapped subarea is contained within the global map.
8 . The device of claim 1 , wherein the quality metric is based on a comparison of the map data of the mapped subarea to an second set of map data that at least partially overlaps with the mapped subarea, wherein the processor is configured to receive the second set of map data from a fixed sensor within the environment.
9 . The device of claim 8 , wherein the fixed sensor comprises a stationary camera, depth camera, infrared camera, or light detection and ranging (LiDAR) sensor.
10 . The device of claim 1 , wherein the processor configured to determine the correspondence comprises the processor configured to determine whether the robot is able to localize itself on the global map.
11 . The device of claim 1 , wherein the quality metric comprises a confidence metric for the map data of the mapped subarea.
12 . The device of claim 1 , wherein the robot is configured to update the global map in real-time when the map data of the mapped subarea is received from the robot.
13 . The device of claim 1 , wherein the processor is configured to transmit the updated global map to the robot.
14 . The device of claim 1 , wherein the processor is further configured to control the robot to use the updated global map for navigation.
15 . A system for map regeneration of an environment, the system comprising:
a map manager configured to maintain a global map of the environment; and
a robot configured to determine a correspondence between a mapped subarea of the environment and a global map of the environment and, based on the correspondence, transmit map data of the mapped subarea to a map manager, wherein the map manager is configured to:
determine a quality metric of the map data of the mapped subarea with respect to the global map; and
update the global map with the map data of the mapped subarea based on the quality metric.
16 . The system of claim 15 , wherein the quality metric is based on a comparison of the map data to a second set of map data that at least partially overlaps with the mapped subarea received from a second robot in the environment.
17 . The system of claim 16 , wherein the map manager is configured to update, based on whether the map data of the mapped subarea satisfies a predefined criterion with respect to the second set of map data, the global map with the map data of the mapped subarea.
18 . The system of claim 17 , wherein the predefined criterion comprises an occupancy metric that characterizes occupied spaces in the mapped subarea and relates to occupancy information of the map data and of the second set of map data that at least partially overlaps with the mapped subarea.
19 . A computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:
determine a correspondence between a global map of an environment and a mapped subarea of the environment, wherein the mapped subarea is received from a robot within the environment;
determine a quality metric of map data of the mapped subarea with respect to the global map; and
update the global map with the map data of the mapped subarea based on the quality metric.
20 . The computer-readable medium of claim 19 , wherein the instructions further cause the one or more processors to determine the quality metric based on a comparison of the map data of the mapped subarea to an second set of map data that at least partially overlaps with the mapped subarea, wherein the second set of map data is received from a second robot in the environment.