TRAINED HUMAN-INTENTION CLASSIFIER FOR SAFE AND EFFICIENT ROBOT NAVIGATION

Number: 20170190051
Published: 2017-07-06
Filed: 2016-01-06
Assignee: DISNEY ENTERPRISES, INC.
Inventors: OSullivan; Carol Ann, Park; Chonhyon, Gilbert; Max L., Ondrej; Jan, Freeman; Kyle G.
CPC: B25J9/1666; G06N7/01; B25J9/163; G06N20/00
Verdict: High Notable software
Source: Google Patents · FreePatentsOnline

The keeper's note

A trained human-intention classifier that helps a mobile robot navigation algorithm compute safe, efficient trajectories around people.

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 thehumans 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.

Background

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic or functional block drawing of a robot control system showing use of a trained human-intention classifier to enhance robot navigation through a workspace shared with mobile entities;

FIG. 2 is an upper perspective view of a workspace being shared by a robot and humans that can be used to gather data for training a human-intention classifier of the present description;

FIG. 3 is a schematic or graph showing results of a training data gather process performed by the inventors for a particular workspace, robot, and group of humans over a predefined time period;

FIG. 4 is a flow diagram of an exemplary process for gathering training data for a human-intention classifier;

FIG. 5 is a schematic or functional block diagram of a motion planner for use in runtime planning of a robot's navigation through a space shared with mobile entities such as a crowd of humans;

FIGS. 6 and 7 illustrate, respectively input feature data and a plot of a trained human-intention classifier; and

FIG. 8 is a graph illustrating the intention-aware repulsive potential curves.DETAILED DESCRIPTION

The following description is directed toward a motion planner with a human-intention classifier, toward robotsincluding such a motion planner (e.g., set of executable code or a software program), andtoward a method of generating the human-intention classifier. In other words, the description addresses problems

Claims

1. A mobile robot, comprising: a drive system operating to move the mobile robot in a workspace; a robot controller transmitting control signals to the drive system tofollow a trajectory through the workspace; a motion planner periodically updating the trajectory; and a human-intention classifier generating a prediction of a behavior for each of a set of mobile entities in the workspace, wherein the motion planner performs the updating of the trajectory using the behavior predictions for the set of mobile entities, and wherein the human-intention classifier is trained to perform the prediction generation in an offline process completed prior to runtime of the mobile robot. 10. A method of creatinga human-intention classifier for a mobile robot, comprising: for a predefined time period, controlling a robot to move along a goal trajectory in a space shared with a plurality of humans; during the predefined time period, recording trajectories of the robot and the humans; and classifying each of the recorded trajectories for the humans as being associated either with blocking of the robot on the goal trajectory or non-blocking of the travel of the robot on the goal trajectory. 15. A method of navigating a robot through a space shared with one or more mobile humans, comprising: positioning a mobile robot in a space; first controlling the mobile robot to move along a first trajectory toward a goal location inthe space; determining a past trajectory of a human in the space; assigning a blocking score to the past trajectory for the human by comparing the past trajectory to a set of prerecorded trajectories each being pre-classified with a probability of blocking behavior; and second controlling the mobile robot to move along either the first trajectory or along asecond trajectory toward the goal location in the space, wherein the second trajectory iscomputed based on the blocking score.