Robot localization and mapping accommodating non-unique landmarks
Robot localization or mapping can be provided without requiring the expense or complexity an “at-a-distance” sensor, such as a camera, a LIDAR sensor, or the like. Adjacency-derived landmark features can be used and non-unique landmark features can be accommodated. Uncertainty in robot pose can be tracked and compared to an adaptive threshold, and non-dock and dock-based localization behavior can be controlled based on the uncertainty, the adaptive threshold, one or more other thresholds, and the accessibility of available differently oriented landmark features, such as perpendicularly oriented straight wall segments landmark features. Available features can be sorted according to a quality metric, and path planning and navigation techniques are also included for helping obtain successful wall-following and localization observations.
1. A mobile cleaning robot, comprising:
a drive system configured to move the robot about a surface of an environment;
a motion sensor configured to monitor movement of the robot; and
a controller circuit, configured to:
track uncertainty in a pose of the robot;
select a localization strategy between a non-dock localization and a dock-based localization based at least in part on a comparison of the tracked uncertainty to one or more threshold values or value ranges including an adaptive first threshold and a second threshold, the selection including selecting the dock-based localization when (i) the adaptive first threshold exceeds the second threshold and (ii) differently oriented non-dock adjacency-derived landmark features are inaccessible for relocalization; and
trigger localization of the robot in the environment using the selected localization strategy.
2. The mobile cleaning robot of claim 1 ,
wherein the controller is configured to:
trigger the non-dock localization using adjacency-derived landmark features when the tracked uncertainty exceeds the adaptive first threshold and is less than the second threshold; and
trigger the non-dock localization using the non-dock adjacency-derived landmark features when (i) the first threshold exceeds the second threshold and (ii) the differently oriented non-dock adjacency-derived landmark features are accessible for the non-dock localization.
3. The mobile cleaning robot of claim 2 , wherein the controller is configured to, when the non-dock localization is successful, adjust the uncertainty and to adjust the first threshold based on the adjusted uncertainty.
4. The mobile cleaning robot of claim 1 , wherein the controller is configured to, when the dock-based localization is successful, adjust the uncertainty and to adjust the first threshold based on the adjusted uncertainty.
5. The mobile cleaning robot of claim 1 , wherein the one or more threshold values or value ranges includes a third threshold greater than the second threshold,
wherein the controller is configured to trigger the dock-based localization when the uncertainty exceeds the third threshold regardless of whether the differently oriented non-dock adjacency-derived landmark features are inaccessible for relocalization.
6. The mobile cleaning robot of claim 1 , wherein the controller is configured to, in response to the tracked uncertainty exceeding the first threshold, determine an accessibility of the differently oriented non-dock adjacency-derived landmark features based at least in part on a then-current robot pose.
7. The mobile cleaning robot of claim 1 , wherein the differently oriented non-dock adjacency-derived landmark features include perpendicular straight wall segments.
8. The mobile cleaning robot of claim 1 , wherein the differently oriented non-dock adjacency-derived landmark features permits non-unique adjacency-derived landmark features that permit at least two different landmark features to have the same length and orientation.
9. The mobile cleaning robot of claim 1 , wherein the controller is configured to trigger the dock-based localization based on a record of past unsuccessful non-dock localizations.
10. The mobile cleaning robot of claim 1 , wherein the controller is configured to plan a path for navigating the robot to an estimated starting location for engaging a non-dock adjacency-derived straight wall segment landmark and substantially maximizing wall-following length during a wall-following that is restricted to a designated lateral side of the robot.
11. The mobile cleaning robot of claim 1 , further comprising a non-imaging and non-distance-ranging adjacency sensor configured to receive adjacency sensor data indicating an obstacle in the environment being adjacent to or in contact with the mobile cleaning robot for use as an adjacency-derived landmark feature in a set of landmark features that allows inclusion of non-unique landmark features including differently oriented straight wall segments,
wherein the controller circuit is configured to trigger localization of the robot further using a landmark feature in the set of landmark features that allows inclusion of the non-unique landmark features.
12. A method of operating a mobile cleaning robot, the method comprising:
moving the robot about a surface of an environment via a drive system of the robot;
monitoring movement of the robot via a sensor data;
tracking uncertainty in at least one of a location or pose of the robot;
selecting a localization strategy between a non-dock localization and a dock-based localization based at least in part on a comparison of the tracked uncertainty to one or more threshold values or value ranges including an adaptive first threshold and a second threshold, wherein selecting the localization strategy includes selecting the dock-based localization when (i) the adaptive first threshold exceeds the second threshold and (ii) differently oriented non-dock adjacency-derived landmark features are inaccessible for relocalization; and
triggering localization of the robot using the selected localization strategy.
13. The method of claim 12 , wherein triggering the localization of the robot includes:
triggering the non-dock localization using adjacency-derived landmark features when the tracked uncertainty exceeds the adaptive first threshold and is less than the second threshold; and
triggering the non-dock localization using the non-dock adjacency-derived landmark features when (i) the first threshold exceeds the second threshold and (ii) the differently oriented non-dock adjacency-derived landmark features are accessible for the non-dock localization.
14. The method of claim 12 , wherein the one or more threshold values or value ranges includes a third threshold greater than the second threshold,
wherein triggering the localization of the robot includes triggering the dock-based localization when the uncertainty exceeds the third threshold regardless of whether the differently oriented non-dock adjacency-derived landmark features are inaccessible for relocalization.
15. The method of claim 12 , further comprising determining an accessibility of the differently oriented non-dock adjacency-derived landmark features based at least in part on a then-current robot pose.
16. The method of claim 12 , comprising receiving adjacency sensor data indicating an obstacle in the environment being adjacent to or in contact with the mobile cleaning robot for use as an adjacency-derived landmark feature in a set of landmark features that allows inclusion of non-unique landmark features including differently oriented straight wall segments,
wherein triggering localization of the robot incudes using a landmark feature in the set of landmark features.
17. The method of claim 12 , further comprising: when at least one of the dock-based localization or the non-dock localization is successful, adjusting the uncertainty, and adjusting the first threshold based on the adjusted uncertainty.
18. The method of claim 12 , wherein the differently oriented non-dock adjacency-derived landmark features include perpendicular straight wall segments.
19. The method of claim 12 , wherein triggering the dock-based localization is further based on a record of past unsuccessful non-dock localizations.
20. The method of claim 12 , further comprising: planning a path for navigating the robot to an estimated starting location for engaging a non-dock adjacency-derived straight wall segment landmark and substantially maximizing wall-following length during a wall-following that is restricted to a designated lateral side of the robot.