Patents by Inventor Akinobu Hayashi
Akinobu Hayashi 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).
-
Patent number: 11958201Abstract: Systems and methods for visuo-tactile object pose estimation are provided. In one embodiment, a method includes receiving image data about an object and receiving depth data about the object. The method also includes generating a visual estimate of the object based on the image data and the depth data. The method further includes receiving tactile data about the object. The method yet further includes generating a tactile estimate of the object based on the tactile data. The method includes estimating a pose of the object based on the visual estimate and the tactile estimate.Type: GrantFiled: September 17, 2020Date of Patent: April 16, 2024Assignee: HONDA MOTOR CO., LTD.Inventors: Nawid Jamali, Huckleberry Febbo, Karankumar Patel, Soshi Iba, Akinobu Hayashi, Itoshi Naramura
-
Publication number: 20240076464Abstract: Polyethylene resin foamed particles according to the present invention are obtained by using, as a base material resin, a non-crosslinked linear low density polyethylene. The linear low density polyethylene is a copolymer of ethylene and an ?-olefin having 8 carbon atoms, and has a melt flow rate and a density in specified ranges. The foamed particle has an average foam size within a specified range, and has a crystal structure that causes an intrinsic peak and a high temperature peak to appear in the first round of a DSC curve obtained under a specific condition. The total fusion heat quantity (?H1) determined from the sum total of a fusion heat quantity (?Hi) of the intrinsic peak and the fusion heat quantity (?Hh) of the high temperature peak is within a specified range.Type: ApplicationFiled: March 3, 2022Publication date: March 7, 2024Inventors: TATSUYA HAYASHI, AKINOBU HIRA
-
Publication number: 20240005206Abstract: A learning device includes a dataset acquisition unit configured to acquire a dataset including state information and action information on which a policy is to be learned, a discrete latent variable estimation unit configured to estimate a discrete latent variable representing characteristics of features from the state information and the action information, an optimal action learning unit configured to learn an optimal action using the state information and the discrete latent variable, a value function estimation unit configured to learn an action value from the state information and the action information, and an identification unit configured to identify a discrete latent variable that maximizes the action value using a result from the optimal action learning unit and a result from the value function estimation unit.Type: ApplicationFiled: June 7, 2023Publication date: January 4, 2024Inventors: Takayuki Osa, Akinobu Hayashi
-
Publication number: 20230234232Abstract: An autonomous control system includes an acquirer configured to acquire state data of a robot, visual data of the robot, and tactile data of the robot and a processor configured to decide on an action of the robot capable of accomplishing a task given to the robot on the basis of the state data, the visual data, and the tactile data. The processor generates first compressed data having a smaller number of dimensions than data obtained by combining the visual data and the tactile data by fusing and dimensionally compressing the visual data and the tactile data. The processor generates second compressed data having a smaller number of dimensions than the tactile data by dimensionally compressing the tactile data. The processor decides on the action on the basis of combined state data obtained by combining the state data, the first compressed data, and the second compressed data into one.Type: ApplicationFiled: January 12, 2023Publication date: July 27, 2023Inventors: Akinobu Hayashi, Itoshi Naramura
-
Publication number: 20220080598Abstract: Systems and methods for visuo-tactile object pose estimation are provided. In one embodiment, a method includes receiving image data about an object and receiving depth data about the object. The method also includes generating a visual estimate of the object based on the image data and the depth data. The method further includes receiving tactile data about the object. The method yet further includes generating a tactile estimate of the object based on the tactile data. The method includes estimating a pose of the object based on the visual estimate and the tactile estimate.Type: ApplicationFiled: September 17, 2020Publication date: March 17, 2022Inventors: Nawid Jamali, Huckleberry Febbo, Karankumar Patel, Soshi Iba, Akinobu Hayashi, Itoshi Naramura
-
Patent number: 10279474Abstract: The invention relates to a method for improving operation of at least one robot. The robot is being operated on the basis of a set of predefined actions. A method comprises generating combined actions by combining at least two actions out of a set of original actions stored in an action library. Storing the combined actions in the actions library in addition to the original actions. Applying a reinforcement learning algorithm to the set of actions stored now in the action library to learn a control policy making use of the original actions and the combined actions. And finally, operating the robot on the basis of the resulting action library.Type: GrantFiled: March 30, 2016Date of Patent: May 7, 2019Assignee: HONDA RESEARCH INSTITUTE EUROPE GMBHInventors: Manuel Mühlig, Michael Gienger, Akinobu Hayashi
-
Publication number: 20160288323Abstract: The invention relates to a method for improving operation of at least one robot. The robot is being operated on the basis of a set of predefined actions. A method comprises generating combined actions by combining at least two actions out of a set of original actions stored in an action library. Storing the combined actions in the actions library in addition to the original actions. Applying a reinforcement learning algorithm to the set of actions stored now in the action library to learn a control policy making use of the original actions and the combined actions. And finally, operating the robot on the basis of the resulting action library.Type: ApplicationFiled: March 30, 2016Publication date: October 6, 2016Inventors: Manuel MÜHLIG, Michael GIENGER, Akinobu HAYASHI
-
Patent number: 9014852Abstract: A system capable of causing an agent to continuously execute a plurality of different subtasks while securing the continuity of behavior of the agent is provided. A plurality of state variable trajectories representing the time series of a state variable of an object are generated according to a stochastic transition model in which the state variable of the object is represented as a random variable. The stochastic transition model is defined so that the transition mode of the state variable is determined according to an execution probability of each subtask in which a probability distribution is represented by a Dirichlet distribution. An operation of the agent is controlled so that the state of the object transits according to one state variable trajectory (desired state variable trajectory) maximizing or optimizing the joint probability of a whole of the stochastic transition model among the plurality of state variable trajectories.Type: GrantFiled: February 22, 2013Date of Patent: April 21, 2015Assignee: Honda Motor Co., Ltd.Inventors: Soshi Iba, Akinobu Hayashi
-
Publication number: 20130345865Abstract: A system capable of causing an agent to continuously execute a plurality of different subtasks while securing the continuity of behavior of the agent is provided. A plurality of state variable trajectories representing the time series of a state variable of an object are generated according to a stochastic transition model in which the state variable of the object is represented as a random variable. The stochastic transition model is defined so that the transition mode of the state variable is determined according to an execution probability of each subtask in which a probability distribution is represented by a Dirichlet distribution. An operation of the agent is controlled so that the state of the object transits according to one state variable trajectory (desired state variable trajectory) maximizing or optimizing the joint probability of a whole of the stochastic transition model among the plurality of state variable trajectories.Type: ApplicationFiled: February 22, 2013Publication date: December 26, 2013Applicant: HONDA MOTOR CO., LTD.Inventors: Soshi Iba, Akinobu Hayashi
-
Patent number: 5678420Abstract: A waste incineration heat conversion system has an incinerator for burning waste discharged from an installation, a heat recovering section for recovering the heat generated by combustion of the waste in the incinerator, a cold generating section for generating cold by utilizing the thus recovered heat, and a pipe line for conveying the thus generated cold to the installation so that the thus conveyed cold is utilized in the installation for a purpose.Type: GrantFiled: October 24, 1995Date of Patent: October 21, 1997Assignee: Hitachi, Ltd.Inventors: Toshiko Fukushima, Yasuo Koseki, Akinobu Hayashi, Ryuichi Kaji