公开(公告)号：WO2022232735A1

公开(公告)日：2022-11-03

申请号：PCT/US2022/071394

申请日：2022-03-28

Applicant: X DEVELOPMENT LLC

Inventor： LAM, Daniel ,  REMBISZ, Justine ,  WEISS, Asa ,  HOLSON, Benjamin

IPC: G05D1/02 ,  G01N35/00 ,  B25J9/00

Abstract: A method includes receiving data collected by at least one sensor on a robotic device, wherein the data is to be used for an ambient environment state representation, and wherein the data represents ambient environment measurements collected at locations of the at least one sensor when the robotic device is passively monitoring an environment such that robotic device navigation is not based on the ambient environment state representation. The method further includes determining the ambient environment state representation using the data collected by the at least one sensor on the robotic device. The method also includes identifying, based on the ambient environment state representation, one or more anomalous ambient environment measurements. The method additionally includes causing, based on the one or more identified anomalous ambient environment measurements, the robotic device to actively monitor the environment such that robotic device navigation is based on the ambient environment state representation.

公开(公告)号：WO2022140709A1

公开(公告)日：2022-06-30

申请号：PCT/US2021/071688

申请日：2021-10-01

Applicant: X DEVELOPMENT LLC

Inventor： REPHAELI, Eden ,  SATAT, Guy ,  LAM, Daniel ,  HOLSON, Benjamin ,  XU, Jiajun

IPC: B25J5/00 ,  B25J9/16 ,  B25J11/00 ,  G06V20/20 ,  G06V40/10

Abstract: A system includes a robotic device, a sensor disposed on the robotic device, and circuitry configured to perform operations. The operations include determining a map that represents stationary features of an environment and receiving, from the sensor, sensor data representing the environment. The operations also include determining, based on the sensor data, a representation of an actor within the environment, where the representation includes keypoints representing corresponding body locations of the actor. The operations also include determining that a portion of a particular stationary feature is positioned within a threshold distance of a particular keypoint and, based on thereon, updating the map to indicate that the portion is to be cleaned. The operations further include, based on the map as updated, causing the robotic device to clean the portion of the particular stationary feature.

公开(公告)号：WO2022197386A1

公开(公告)日：2022-09-22

申请号：PCT/US2022/015252

申请日：2022-02-04

Applicant: X DEVELOPMENT LLC

Inventor： KHANSARI ZADEH, Seyed Mohammad ,  JANG, Eric ,  LAM, Daniel ,  KAPPLER, Daniel ,  BENNICE, Matthew ,  AUSTIN, Brent ,  BAI, Yunfei ,  LEVINE, Sergey ,  IRPAN, Alexander ,  SIEVERS, Nicolas ,  FINN, Chelsea

IPC: B25J9/16

Abstract: Implementations described herein relate to training and refining robotic control policies using imitation learning techniques. A robotic control policy can be initially trained based on human demonstrations of various robotic tasks. Further, the robotic control policy can be refined based on human interventions while a robot is performing a robotic task. In some implementations, the robotic control policy may determine whether the robot will fail in performance of the robotic task, and prompt a human to intervene in performance of the robotic task. In additional or alternative implementations, a representation of the sequence of actions can be visually rendered for presentation to the human can proactively intervene in performance of the robotic task.