Systems and methods for performing occlusion detection
First Claim
Patent Images
1. A mobile robot configured to navigate an operating environment, comprising:
- a body containing;
a drive configured to translate the mobile robot in a direction of motion;
a machine vision system comprising a camera that captures images of the operating environment of the mobile robot;
a processor;
memory containing a simultaneous localization and mapping (SLAM) application and a behavioral control application;
wherein the behavioral control application directs the processor to;
capture images using the machine vision system;
detect a presence of an occlusion obstructing a portion of a field of view of the camera based on the captured images; and
generate a notification when the occlusion obstructing the portion of the field of view of the camera is detected; and
wherein the behavioral control application further directs the processor to maintain occlusion detection data describing occluded and unobstructed portions of images being used by the SLAM application.
4 Assignments
0 Petitions
Accused Products
Abstract
The present invention provides a mobile robot configured to navigate an operating environment, that includes a machine vision system comprising a camera that captures images of the operating environment using a machine vision system; detects the presence of an occlusion obstructing a portion of the field of view of a camera based on the captured images, and generate a notification when an occlusion obstructing the portion of the field of view of the camera is detected, and maintain occlusion detection data describing occluded and unobstructed portions of images being used by the SLAM application.
118 Citations
20 Claims
-
1. A mobile robot configured to navigate an operating environment, comprising:
a body containing; a drive configured to translate the mobile robot in a direction of motion; a machine vision system comprising a camera that captures images of the operating environment of the mobile robot; a processor; memory containing a simultaneous localization and mapping (SLAM) application and a behavioral control application; wherein the behavioral control application directs the processor to; capture images using the machine vision system; detect a presence of an occlusion obstructing a portion of a field of view of the camera based on the captured images; and generate a notification when the occlusion obstructing the portion of the field of view of the camera is detected; and wherein the behavioral control application further directs the processor to maintain occlusion detection data describing occluded and unobstructed portions of images being used by the SLAM application. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
19. A mobile robot configured to navigate an operating environment, comprising:
-
a body containing; a drive mechanism configured to translate the mobile robot in a direction of motion; an odometry sensor system; a machine vision sensor system comprising a camera that captures images of the operating environment of the mobile robot; and at least one processor; memory containing a behavioral control application and a visual simultaneous localization and mapping (VSLAM) application, a pre-loaded map of the operating environment, and an estimated robot pose; wherein the behavioral control application directs a processor to; actuate the drive mechanism and capture; odometry data using the odometry sensor system; and images using the machine vision sensor system; determine updated robot poses and update the pre-loaded map of the operating environment by providing at least the captured images and the odometry data as inputs to the VSLAM application; update occlusion detection data describing portions of images being used by the VSLAM application to determine the updated robot poses and update the pre-loaded map of the operating environment; detect a presence of an occlusion obstructing the camera based on the occlusion detection data; and provide a notification when the occlusion detection data indicates the presence of the occlusion obstructing the camera, wherein the VSLAM application directs a processor to; estimate a location of the mobile robot within the pre-loaded map of the environment based upon the images captured by the machine vision sensor system and the odometry data captured by the odometry sensor system, and update the pre-loaded map of the operating environment based upon the images captured by the machine vision sensor system. - View Dependent Claims (20)
-
Specification