Agility Robotics, Inc.

Shelby Cass

Brent Griffin

Robert Debortoli

A method in accordance with at least some embodiments of the present technology includes providing an image and a plurality of indicators. The image depicts an environment of a robot and corresponds to vision data collected by a vision sensor of the robot at a given timestep of a plurality of timesteps. Indicators among the plurality of indicators correspond to respective expected object locations within the environment. The method further includes receiving user input indicating a user identified state of the given one of the expected object locations based at least partially on the image. Finally, the method includes associating additional vision data with the user identified state of the given one of the expected object locations at least partially in response to receiving the user input. The additional vision data is collected by the vision sensor at additional timesteps of the plurality of timesteps.

Growing community of inventors

Corvallis, OR, United States of America

Shelby Cass