公开(公告)号：US20170190051A1

公开(公告)日：2017-07-06

申请号：US14989345

申请日：2016-01-06

Applicant: DISNEY ENTERPRISES, INC.

Inventor： CAROL ANN O'SULLIVAN ,  CHONHYON PARK ,  MAX L. GILBERT ,  JAN ONDREJ ,  KYLE G. FREEMAN

IPC: B25J9/16

CPC classification number: B25J9/1666 ,  B25J9/163 ,  G05D1/0088 ,  G05D2201/0211 ,  G06N3/00 ,  Y10S901/01 ,  Y10S901/46 ,  Y10S901/49

Abstract: A trained classifier to be used with a navigation algorithm for use with mobile robots to compute safe and efficient trajectories. An offline learning process is used to train a classifier for the navigation algorithm (or motion planner), and the classifier functions, after training is complete, to accurately detect intentions of humans within a space shared with the robot to block the robot from traveling along its current trajectory. At runtime, the trained classifier can be used with regression based on past trajectories of humans (or other tracked, mobile entities) to predict where the humans will move in the future and whether the humans are likely to be blockers. The planning algorithm or motion planner generates trajectories based on predictions of human behavior that allow the robot to navigate amongst crowds of people more safely and efficiently.

公开(公告)号：US20170305013A1

公开(公告)日：2017-10-26

申请号：US15137735

申请日：2016-04-25

Applicant: DISNEY ENTERPRISES, INC.

Inventor： KYLE G. FREEMAN

IPC: B25J9/16

CPC classification number: B25J9/1664 ,  B25J9/1697 ,  G01C21/00 ,  G05D1/024 ,  G05D1/0274 ,  G05D1/0278 ,  Y10S901/01 ,  Y10S901/47

Abstract: A method for localizing robots and other objects when ranging devices are partially obstructed. The method includes retrieving a digital map of the space and then identifying potential locations for the object in the space. The method involves generating a prediction of expected ranges between the object and surfaces in the space at each of the potential locations. The method includes operating a ranging device to measure ranges to the surfaces in the space and comparing, for each of the predictions, the measured ranges with the expected ranges to identify the most accurate prediction. The comparing includes weighting ranges that are too small neutrally as these are strikes on obstructions, weighting ranges that provide matches positively, and weighting ranges that are too large negatively, and the comparing then involves summing the weighted ranges to identify a most accurate one of the predictions and associated current location for the object.