Nara Institute of Science and Technology

Tsukasa Ogasawara

Jun Ueda

Ming Ding

Yutaka Ishida

Takuya Kiyokawa

Yuichi Kurita

Atsutoshi Ikeda

Keita Tomochika

Jun Takamatsu

Yasuhiro Ono

Provided are a method and a device that can efficiently generate a training dataset. Object information is associated with a visual marker, a training dataset generation jig that is configured from a base part and a marker is used, said base part being provided with an area that serves as a guide for positioning a target object and said marker being fixed on the base part, the target object is positioned using the area as a guide and in this condition an image group of the entire object including the marker is acquired, the object information that was associated with the visual marker is acquired from the acquired image group, a reconfigured image group is generated from this image group by performing a concealment process on a region corresponding to the visual marker or the training dataset generation jig, a bounding box is set in the reconfigured image group on the basis of the acquired object information, information relating to the bounding box, the object information, and estimated target object position information and posture information are associated with a captured image, and a training dataset for performing object recognition and position/posture estimation for the target object is generated.

Learning dataset creation method and device

A robot hand which can be formed in the shape of a human hand and improves maintenance performance, and generates fingertip force is provided. A robot hand () of the present invention is provided with a hand section () and an arm section (). The hand section () corresponds to the portion of a human hand which is forward from the wrist thereof. The arm section () is provided with a drive device (d) including a plurality of motors and generates driving torque for driving each movable section of the hand section (). The hand section () and the arm section () are configured in such a manner that a group (g) of wrist-mounted pulleys which are first driving torque transmitting members coaxially arranged and a group (r) of arm pulleys which are second driving torque transmitting members coaxially arranged transmit driving torque generated by the drive device (d) to each movable section of the hand section () via gears meshing with each other.

Robot hand

In one embodiment of the present invention, a driving force calculating device calculates a driving force so that a desired load is applied on a muscle while a power assisting device assists and/or resists a predetermined motion. A driving force calculating section of a driving force calculating device calculates a driving force of each driving section according to a joint torque table on joint torque necessary for a rotational motion and an assigned muscle force table on an assigned muscle force input to a muscle force input section. For the calculation is used a muscle/artificial muscle integrated model data integrally including musculoskeletal model data on a musculoskeletal model of a human and artificial muscle model data on an artificial muscle model of the power assisting device. The driving force is calculated through optimization using Crowninshield performance function under constraint conditions of the joint torque and the assigned muscle force.

Driving force calculating device, driving force calculating method, power assisting device, program, and computer-readable storage medium

A tactile sensor of an embodiment of the present invention includes: a sensing section having an elastic member at a portion which contacts a measurement target; an image acquiring section for acquiring as image information the state of a contact surface of the measurement target and the elastic member, before and after application of an external force tangential to the contact surface; a deformation analyzing section for analyzing deformation information of the contact surface, based on the image information acquired by the image acquiring section; an external force detecting section for detecting the external force applied tangential to the contact surface; and an estimating section for estimating a slippage margin between the measurement target and the elastic member, based on (I) the deformation information of the contact surface, which information acquired by the deformation analyzing section, (II) the external force detected by the external force detecting section, and (III) an object constant of the elastic member. With this tactile sensor, slippage margin can be simply but accurately estimated.

Tactile sensor and use thereof

A driving force calculating device () includes: a setting information acquiring section () which acquires (a) muscle specifying information that specifies a target muscle, which is a muscle whose function is assisted or resisted by a driving section and (b) a target value of muscle force to be produced by the target muscle at the time of rotational motion under assistance or resistance of the driving section; and a target muscle force evaluating section () which acquires an estimated muscle force table () and musculoskeletal model data () and evaluates the feasibility of the target value of the target muscle. With this arrangement, it is possible to reduce time required to calculate a driving force of the driving section of the power assisting device.

Driving force calculating device, driving force calculating method, power assisting device, driving force calculating program, and computer-readable storage medium

Growing community of inventors

Ikoma, Japan

Tsukasa Ogasawara