Patents by Inventor Yuwei Cui
Yuwei Cui has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230111841Abstract: Embodiments relate to a processing node in a temporal memory system that performs temporal pooling or processing by activating cells where the activation of a cell is maintained longer if the activation of the cell were previously predicted or activation on more than a certain portion of associated cells in a lower node was correctly predicted. An active cell correctly predicted to be activated or an active cell having connections to lower node active cells that were correctly predicted to become active contribute to accurate prediction, and hence, is maintained active longer than cells activated but were not previously predicted to become active. Embodiments also relate to a temporal memory system for detecting, learning, and predicting spatial patterns and temporal sequences in input data by using action information.Type: ApplicationFiled: November 18, 2022Publication date: April 13, 2023Inventors: Jeffrey C. Hawkins, Subutai Ahmad, Yuwei Cui, Chetan Surpur
-
Patent number: 11537922Abstract: Embodiments relate to a processing node in a temporal memory system that performs temporal pooling or processing by activating cells where the activation of a cell is maintained longer if the activation of the cell were previously predicted or activation on more than a certain portion of associated cells in a lower node was correctly predicted. An active cell correctly predicted to be activated or an active cell having connections to lower node active cells that were correctly predicted to become active contribute to accurate prediction, and hence, is maintained active longer than cells activated but were not previously predicted to become active. Embodiments also relate to a temporal memory system for detecting, learning, and predicting spatial patterns and temporal sequences in input data by using action information.Type: GrantFiled: April 26, 2019Date of Patent: December 27, 2022Assignee: Numenta, Inc.Inventors: Jeffrey C. Hawkins, Subutai Ahmad, Yuwei Cui, Chetan Surpur
-
Patent number: 11320834Abstract: The present application provides methods and systems for mapping, localization, navigation and control and a mobile robot. Through the technical solution of acquiring a first projected image and second projected image of an entity object when the mobile robot moves to a first position and a second position respectively, building a map or localizing the mobile robot based on angle information of the entity object relative to the mobile robot at the first position and at the second position, the first position and the second position, planning a navigation route based on the built map and localization, and controlling the mobile robot to move along the navigation route, position information of the entity object can be determined precisely, and precision of the built map can be improved. Moreover, movement of the mobile robot can be controlled precisely, and operating precision of the mobile robot and human-computer interaction can be improved.Type: GrantFiled: August 6, 2019Date of Patent: May 3, 2022Assignees: ANKOBOT (SHENZHEN) SMART TECHNOLOGIES CO., LTD., ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Chongxing Li
-
Publication number: 20220057212Abstract: The present application provides a method for updating a map and a mobile robot. The method comprises: acquiring a current map and a corresponding current location data set constructed when a first movement device performs a navigation and movement operation in a physical space, wherein the first movement device performs the navigation and movement operation based on a reference map and a corresponding reference location data set which are corresponding to the physical space and stored in advance; performing a data fusion process on the reference map and the corresponding reference location data set as well as the current map and the corresponding current location data set; taking a map and a corresponding location data set after being data fusion processed as a new reference map and a corresponding new reference location data set in the first movement device. The present application provides a solution for persistence of a map.Type: ApplicationFiled: November 5, 2021Publication date: February 24, 2022Inventors: Yuwei CUI, Wei LI
-
Patent number: 11204247Abstract: The present application provides a method for updating a map and a mobile robot. The method comprises: acquiring a current map and a corresponding current location data set constructed when a first movement device performs a navigation and movement operation in a physical space, wherein the first movement device performs the navigation and movement operation based on a reference map and a corresponding reference location data set which are corresponding to the physical space and stored in advance; performing a data fusion process on the reference map and the corresponding reference location data set as well as the current map and the corresponding current location data set; taking a map and a corresponding location data set after being data fusion processed as a new reference map and a corresponding new reference location data set in the first movement device. The present application provides a solution for persistence of a map.Type: GrantFiled: October 24, 2019Date of Patent: December 21, 2021Assignees: ANKOBOT (SHENZHEN) SMART TECHNOLOGIES CO., LTD., ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Wei Li
-
Patent number: 11130238Abstract: The present application provides a method for controlling a movement of a mobile robot, a mobile robot and a computer storage medium. The method for controlling a movement of a mobile robot, the mobile robot is provided with an image acquisition device, and the method comprises the following steps: identifying a border line between a first obstacle and a ground from an image captured by the image acquisition device; determining a first relative position between a physical position corresponding to the border line and the mobile robot based on preset physical reference information; and controlling the movement of the mobile robot based on the first relative position. Through the technical solution disclosed in the present application, the localization accuracy of the mobile robot can be improved.Type: GrantFiled: December 8, 2020Date of Patent: September 28, 2021Assignees: ANKOBOT (SHENZHEN) SMART TECHNOLOGIES CO., LTD., ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Chongxing Li, Zijing Wang
-
Publication number: 20210201181Abstract: Embodiments relate to performing inference, such as object recognition, based on sensory inputs received from sensors and location information associated with the sensory inputs. The sensory inputs describe one or more features of the objects. The location information describes known or potential locations of the sensors generating the sensory inputs. An inference system learns representations of objects by characterizing a plurality of feature-location representations of the objects, and then performs inference by identifying or updating candidate objects consistent with feature-location representations observed from the sensory input data and location information. In one instance, the inference system learns representations of objects for each sensor. The set of candidate objects for each sensor is updated to those consistent with candidate objects for other sensors, as well as the observed feature-location representations for the sensor.Type: ApplicationFiled: March 11, 2021Publication date: July 1, 2021Inventors: Jeffrey C. Hawkins, Subutai Ahmad, Yuwei Cui, Marcus Anthony Lewis
-
Publication number: 20210201509Abstract: The present application provides a monitoring method and device for a mobile target, a monitoring system and mobile robot. The present application through the technical solution that acquiring multiple-frame images captured by an image acquisition device under an moving state of a robot in a monitored region, selecting at least two-frame images with an overlapped region from the multiple-frame images, performing comparison between the selected images by image compensation method or feature matching method, and outputting monitoring information containing a mobile target which moves relative to a static target based on the result of comparison, wherein the position of the mobile target has an attribute of indefinite change, the mobile target in the monitored region can be recognized precisely during movement of the mobile robot, and monitoring information about the mobile target can be generated to prompt correspondingly, thereby safety of the monitored region can be effectively ensured.Type: ApplicationFiled: February 25, 2021Publication date: July 1, 2021Inventor: Yuwei CUI
-
Patent number: 11042760Abstract: The present application provides a mobile robot, a control method and control system of the mobile robot. The control method comprises the following steps: acquiring a depth image obtained through photography of the at least one image acquisition device; wherein, the depth image contains a depth image of the target obstacle; identifying an obstacle type of the target obstacle from the depth image; controlling a navigation movement and/or behavior of the mobile robot based on an acquired positional relationship between the target obstacle and the mobile robot, and the identified obstacle type. In the present application, the type of the target obstacle can be detected effectively, and the behavior of the mobile robot can be controlled based on the detection result correspondingly.Type: GrantFiled: May 29, 2020Date of Patent: June 22, 2021Assignee: ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Chongxing Li
-
Publication number: 20210182579Abstract: The present application provides a mobile robot, a control method and control system of the mobile robot. The control method comprises the following steps: acquiring a depth image obtained through photography of the at least one image acquisition device; wherein, the depth image contains a depth image of the target obstacle; identifying an obstacle type of the target obstacle from the depth image; controlling a navigation movement and/or behavior of the mobile robot based on an acquired positional relationship between the target obstacle and the mobile robot, and the identified obstacle type. In the present application, the type of the target obstacle can be detected effectively, and the behavior of the mobile robot can be controlled based on the detection result correspondingly.Type: ApplicationFiled: February 22, 2021Publication date: June 17, 2021Inventors: Yuwei CUI, Chongxing LI
-
Publication number: 20210138657Abstract: The present application provides a method for controlling a movement of a mobile robot, a mobile robot and a computer storage medium. The method for controlling a movement of a mobile robot, the mobile robot is provided with an image acquisition device, and the method comprises the following steps: identifying a border line between a first obstacle and a ground from an image captured by the image acquisition device; determining a first relative position between a physical position corresponding to the border line and the mobile robot based on preset physical reference information; and controlling the movement of the mobile robot based on the first relative position. Through the technical solution disclosed in the present application, the localization accuracy of the mobile robot can be improved.Type: ApplicationFiled: December 8, 2020Publication date: May 13, 2021Inventors: Yuwei CUI, Chongxing LI, Zijing WANG
-
Patent number: 10977566Abstract: Embodiments relate to performing inference, such as object recognition, based on sensory inputs received from sensors and location information associated with the sensory inputs. The sensory inputs describe one or more features of the objects. The location information describes known or potential locations of the sensors generating the sensory inputs. An inference system learns representations of objects by characterizing a plurality of feature-location representations of the objects, and then performs inference by identifying or updating candidate objects consistent with feature-location representations observed from the sensory input data and location information. In one instance, the inference system learns representations of objects for each sensor. The set of candidate objects for each sensor is updated to those consistent with candidate objects for other sensors, as well as the observed feature-location representations for the sensor.Type: GrantFiled: May 12, 2017Date of Patent: April 13, 2021Assignee: Numenta, Inc.Inventors: Jeffrey C. Hawkins, Subutai Ahmad, Yuwei Cui, Marcus Anthony Lewis
-
Patent number: 10970859Abstract: The present application provides a monitoring method and device for a mobile target, a monitoring system and mobile robot. The present application through the technical solution that acquiring multiple-frame images captured by an image acquisition device under an moving state of a robot in a monitored region, selecting at least two-frame images with an overlapped region from the multiple-frame images, performing comparison between the selected images by image compensation method or feature matching method, and outputting monitoring information containing a mobile target which moves relative to a static target based on the result of comparison, wherein the position of the mobile target has an attribute of indefinite change, the mobile target in the monitored region can be recognized precisely during movement of the mobile robot, and monitoring information about the mobile target can be generated to prompt correspondingly, thereby safety of the monitored region can be effectively ensured.Type: GrantFiled: July 26, 2019Date of Patent: April 6, 2021Assignees: ANKOBOT (SHENZHEN) SMART TECHNOLOGIES CO., LTD., ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventor: Yuwei Cui
-
Publication number: 20210049376Abstract: The present application provides a mobile robot, a control method and control system of the mobile robot. The control method comprises the following steps: acquiring a depth image obtained through photography of the at least one image acquisition device; wherein, the depth image contains a depth image of the target obstacle; identifying an obstacle type of the target obstacle from the depth image; controlling a navigation movement and/or behavior of the mobile robot based on an acquired positional relationship between the target obstacle and the mobile robot, and the identified obstacle type. In the present application, the type of the target obstacle can be detected effectively, and the behavior of the mobile robot can be controlled based on the detection result correspondingly.Type: ApplicationFiled: May 29, 2020Publication date: February 18, 2021Inventors: Yuwei CUI, Chongxing LI
-
Publication number: 20200356582Abstract: The present application provides a method for updating a map and a mobile robot. The method comprises: acquiring a current map and a corresponding current location data set constructed when a first movement device performs a navigation and movement operation in a physical space, wherein the first movement device performs the navigation and movement operation based on a reference map and a corresponding reference location data set which are corresponding to the physical space and stored in advance; performing a data fusion process on the reference map and the corresponding reference location data set as well as the current map and the corresponding current location data set; taking a map and a corresponding location data set after being data fusion processed as a new reference map and a corresponding new reference location data set in the first movement device. The present application provides a solution for persistence of a map.Type: ApplicationFiled: October 24, 2019Publication date: November 12, 2020Inventors: Yuwei CUI, Wei LI
-
Patent number: 10776687Abstract: Embodiments relate to a processing node of a hierarchical temporal memory (HTM) system with a union processor that enables a more stable representation of sequences by unionizing or pooling patterns of a temporal sequence. The union processor biases the HTM system so a learned temporal sequence may be more quickly recognized. The union processor includes union elements that are associated with incoming spatial patterns or with cells that represent temporal relationships between the spatial patterns. A union element of the union processor may be activated if a persistence score of the union element satisfies a predetermined criterion. The persistence score of the detector is updated based on the activation states of the spatial patterns or cells associated with the detector. After activation, the union element remains active for a period longer than a time step for performing the spatial pooling.Type: GrantFiled: March 3, 2016Date of Patent: September 15, 2020Assignee: Numenta, Inc.Inventors: Jeffrey C. Hawkins, Yuwei Cui
-
Publication number: 20200184658Abstract: The present application provides a monitoring method and device for a mobile target, a monitoring system and mobile robot. The present application through the technical solution that acquiring multiple-frame images captured by an image acquisition device under an moving state of a robot in a monitored region, selecting at least two-frame images with an overlapped region from the multiple-frame images, performing comparison between the selected images by image compensation method or feature matching method, and outputting monitoring information containing a mobile target which moves relative to a static target based on the result of comparison, wherein the position of the mobile target has an attribute of indefinite change, the mobile target in the monitored region can be recognized precisely during movement of the mobile robot, and monitoring information about the mobile target can be generated to prompt correspondingly, thereby safety of the monitored region can be effectively ensured.Type: ApplicationFiled: July 26, 2019Publication date: June 11, 2020Inventor: Yuwei CUI
-
Publication number: 20200159246Abstract: The present application provides methods and systems for mapping, localization, navigation and control and a mobile robot. Through the technical solution of acquiring a first projected image and second projected image of an entity object when the mobile robot moves to a first position and a second position respectively, building a map or localizing the mobile robot based on angle information of the entity object relative to the mobile robot at the first position and at the second position, the first position and the second position, planning a navigation route based on the built map and localization, and controlling the mobile robot to move along the navigation route, position information of the entity object can be determined precisely, and precision of the built map can be improved. Moreover, movement of the mobile robot can be controlled precisely, and operating precision of the mobile robot and human-computer interaction can be improved.Type: ApplicationFiled: August 6, 2019Publication date: May 21, 2020Inventors: Yuwei Cui, Chongxing Li
-
Patent number: 10545497Abstract: The present application provides a control method and device for a mobile robot, a mobile robot. The present application through the technical solution that identifying images captured by a visual sensing device and when determining there is a human body in a spatial range taken by the visual sensing device, determining a pose instruction provided by the human body according to at least one image of the human body, and controlling the robot to perform a corresponding operation based on the pose instruction, the pose instruction of the human can be accurately identified, and accuracy and flexibility in executing instruction by the robot can be improved. Meanwhile, the robot can be controlled in real time according to the pose instruction, such that interactivity between the robot and a user can be improved, humanized operation of the robot can be improved, and user experience can be enhanced.Type: GrantFiled: March 25, 2019Date of Patent: January 28, 2020Assignee: ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Zijing Wang
-
Patent number: 10518414Abstract: The present application provides a navigation method, a navigation system, a movement control system and a mobile robot. The navigation method comprises the following steps: identifying a target object corresponding to a preset object type label from images captured by the image acquisition device; determining the relative position between the target and the mobile robot based on at least one image which contains the identified target and is captured by the image acquisition device; and generating a map marked with the object type label through marking corresponding preset object type label on a preset map based on the relative position, wherein the map is used for navigating the mobile robot. In the present application, mobile robot can move to the destination position after the destination position contained in the user instruction being identified based on the object type labels on the map, so as to improve human-computer interactivity.Type: GrantFiled: September 10, 2018Date of Patent: December 31, 2019Assignees: ANKOBOT (SHENZHEN) SMART TECHNOLOGIES CO., LTD., ANKOBOT (SHANGHAI) SMART TECHNOLOGIES CO., LTD.Inventors: Yuwei Cui, Chongxing Li, Renhua Wen, Zijing Wang