초록 ▼

A method of localizing a mobile robot includes receiving sensor data of a scene about the robot and executing a particle filter having a set of particles. Each particle has associated maps representing a robot location hypothesis. The method further includes updating the maps associated with each pa

A method of localizing a mobile robot includes receiving sensor data of a scene about the robot and executing a particle filter having a set of particles. Each particle has associated maps representing a robot location hypothesis. The method further includes updating the maps associated with each particle based on the received sensor data, assessing a weight for each particle based on the received sensor data, selecting a particle based on its weight, and determining a location of the robot based on the selected particle.

대표청구항 ▼

1. A method comprising: maneuvering a robot about a scene;emitting light onto the scene;capturing images of the scene using a depth-perceptive imaging sensor, the images comprising an active illumination image and an ambient illumination image, each image comprising three-dimensional depth data and

1. A method comprising: maneuvering a robot about a scene;emitting light onto the scene;capturing images of the scene using a depth-perceptive imaging sensor, the images comprising an active illumination image and an ambient illumination image, each image comprising three-dimensional depth data and brightness data;executing a particle filter having a set of particles, each particle having an associated occupancy grid map, an associated feature map, and a robot location hypothesis;updating the occupancy grid map associated with each particle based on the images;for each image: instantiating an image pyramid comprising a set of scaled images, each scaled image having a scale relative to the image;identifying at least one feature point in the scaled images; andupdating the corresponding feature map of each particle with the identified at least one feature point;determining a location of an object in the scene based on the images and at least one particle of the particle filter;assigning a confidence level for the location of the object based on the three-dimensional depth data and the brightness data of the images; andmaneuvering the robot in the scene based on the location of the object and the corresponding confidence level. 2. The method of claim 1, further comprising: receiving a proximity indication of the robot being near the object from a proximity sensor of the robot; andin response to the proximity indication, increasing an image capture rate of the depth-perceptive imaging sensor. 3. The method of claim 1, further comprising constructing an object occupancy map of the scene based on the images. 4. The method of claim 1, further comprising maneuvering the robot to avoid the object. 5. The method of claim 1, further comprising: assessing a weight for each particle based on the images;selecting at least one particle based on its weight; anddetermining a location of the robot based on the at least one selected particle. 6. The method of claim 1, wherein the three-dimensional depth data of each image comprises a three-dimensional point cloud. 7. The method of claim 6, further comprising accumulating cloud points of the three-dimensional point cloud of each image in cells of the occupancy grid map based on first and second coordinates of the cloud points, each cell accumulating a height variance based on a third coordinate of the cloud points. 8. The method of claim 1, further comprising calculating a Harris Corner Score to identify feature points associated with a corner feature of the scene. 9. The method of claim 8, further comprising selecting feature points as candidate feature points that have at least one of a local maximum Harris Corner Score or a Harris Corner Score substantially equal to the local maximum Harris Corner Score in a threshold area. 10. The method of claim 9, further comprising selecting feature points as candidate feature points that have at least one of a local maximum Harris Corner Score or a Harris Corner Score within about 20% of the local maximum Harris Corner Score within a 10 pixel radius of the feature point having the local maximum Harris Corner Score. 11. The method of claim 1, further comprising selecting a feature point of a scaled image as a key point for tracking and producing a descriptor of that key point. 12. The method of claim 11, further comprising identifying the key point in a subsequent image using the descriptor. 13. The method of claim 11, wherein the descriptor comprises feature points within a threshold distance of the key point on the corresponding scaled image of the key point. 14. The method of claim 13, further comprising: sampling feature points of the descriptor;recording a brightness level for each feature point; andnormalizing the brightness levels to have a mean of zero and a variance of one. 15. The method of claim 14, further comprising, before sampling the feature points, at least one of: blurring the scaled image before sampling the feature points; orrotating the feature points by a threshold angle. 16. The method of claim 14, further comprising sampling feature points within a threshold area of the scaled image about the key point. 17. The method of claim 14, further comprising: producing a descriptor for each feature point of a set of feature points;comparing feature descriptors of first and second images to identify a common feature point; andsearching respective image pyramids of the first and second images to find the common feature point. 18. The method of claim 17, further comprising searching within a threshold area of the scaled images of the images pyramids for the common feature point. 19. The method of claim 17, further comprising determining the threshold area based on at least one of a known previous feature point location or a robot drive trajectory.