Abstract: During movement control wherein a moving body tracks a user while in front of the user, a coordinate obtained from the torso of the user is used as the X coordinate of the position of the user, and among the position of the body of the user and the position of the feet of the user, the position which is closer to the moving body is used as the Y coordinate. Thus, the moving body is able to move to a target position in the left/right direction (the X coordinate) and the front/back direction (the Y coordinate) that is appropriate with respect to the user. Accordingly, the distance between the moving body and the user can be maintained appropriately, and movement control in front of the user and that is not an impediment to the user can be accomplished.

Abstract: An image recognition device executes a Hilbert scan of frame image data constituting moving-image data to generate one-dimensional spatial image data, and further arrays the one-dimensional spatial image data in a time direction to generate two-dimensional spatio-temporal image data that holds spatial information and temporal information. The image recognition device converts the moving-image data into the two-dimensional spatio-temporal image data while holding the spatial and temporal information. By means of a CNN unit, the image recognition device executes a convolution process wherein a two-dimensional filter is used on the spatio-temporal image data to image-recognize a behavior of a pedestrian who is a recognition object.

Abstract: A vehicle includes: a vehicle body; one or more turn wheels turnable to right and left; a lean actuator configured to apply a lean torque to the vehicle body; a turn wheel support unit; and a controller. The turn wheel support unit includes: a supporting member that rotatably supports the one or more turn wheels; and a turning actuator configured to apply a turning torque to the supporting member. The controller is configured to determine a target lean angle, and determine the target lean torque and the target turning torque using the target lean angle.

Abstract: An invention to measure the blood pressure of a living body in a noncontact and non-invasive manner. An experimental device irradiates a subject with microwaves from a transmission antenna and a reception antenna. The phase of the reflected wave varies with micro-movements caused by the heartbeat, a pulse wave is detected from the difference between the emitted wave and the reflected wave. The experimental device detects the blood pressure from a blood pressure sensor mounted on the subject while detecting the pulse wave. A spectrum of the pulse wave of is generated by Fourier transformation. A comparison of the spectrum and the blood pressure of the subject detected by the blood pressure sensor reveals a positive correlation of the systolic blood pressure with an eighth harmonic wave from the pulse wave. A device can detect a pulse wave in a non-invasive manner using a microwave and measures systolic blood pressure.

Abstract: A vehicle includes: a vehicle body; one or more turn wheels turnable to right and left; turn wheel support unit; and controller. The turn wheel support unit includes: a supporting member that rotatably supports the one or more turn wheels; and a turning actuator configured to apply a turning torque to the supporting member. The controller is configured to determine a target lean angle, determine a first type control value by using a difference between the target lean angle and a lean angle, and determine the target turning torque by using one or more control values including the first type control value. Where the first type control value indicates a torque that causes the supporting member to turn in a direction opposite to a target direction that is a rotational direction to rotate the vehicle body in its width direction so that the lean angle approaches the target lean angle.

Abstract: A hardware configuration is constructed for calculating at high speed the co-occurrence of luminance gradient directions between differing resolutions for a subject image. In an image processing device, a processing line for high-resolution images, a processing line for medium-resolution images, and a processing line for low-resolution images are arranged in parallel, and the luminance gradient directions are extracted for each pixel simultaneously in parallel from images having the three resolutions. Co-occurrence matrix preparation units prepare co-occurrence matrices by using the luminance gradient directions extracted from these images having the three resolutions, and a histogram preparation unit outputs a histogram as an MRCoHOG feature amount by using these matrices. To concurrently processing the images having the three resolutions, high-speed processing can be performed, and moving pictures output from a camera can be processed in real time.

Abstract: A vehicle includes: a vehicle body; lean mechanism; operation input, lean control, and turn wheel support units. The turn wheel support unit includes: a supporting member supporting one or more turn wheels; turning actuator applying to the supporting member a torque for turning the supporting member; and turn control unit using a control parameter to control a turning actuator torque. The turn control unit includes: a specifying module using the control parameter to specify one or more turn wheels target direction; first determination module determining a first control value for causing the one or more turn wheels direction to approach the target direction; an actuation control value determination module using the first control value to determine an actuation control value; and torque control module controlling a turning actuator torque according to the actuation control value. The first determination module uses the vehicle velocity to adjust the first control value.

Abstract: An image recognition device sets an overall observation region which surrounds a whole body of an object and partial observation regions which surround characteristic parts of the object respectively to locations in an image which are estimated to include captured images of the object. The device clips images in the overall observation region and the partial observation regions, and calculates similarity degrees between them and previously learned images on the basis of a combination of two image feature amounts. The device calculates an optimum ratio in combining the HOG feature amount and the color distribution feature amount individually for the regions. This ratio is determined by setting a weight parameter ?i for setting a weight used for combining the HOG feature amount and the color distribution feature amount to be included in a state vector and subjecting the result to complete search by a particle filter.

Abstract: A vehicle includes: a vehicle body; three or more wheels; an operation input unit to be operated to input a turning direction; and a lean mechanism for leaning the vehicle body in its width direction. Within at least partial range of vehicle velocity, the vehicle is configured to travel in a mode in which the vehicle body is leaned by the lean mechanism according to an input into the operation input unit, and a steering angle of a steered wheel changes following a lean of the vehicle body. And, the vehicle includes a changing device for changing a turn resistance force acting between the vehicle body and the steered wheel.

Abstract: An image reproduction device includes: a reflection hologram which reflects reproduction illumination light emitted from a first side to reproduce a reproduced image on the first side; and a beam splitting element which is arranged on the first side of the reflection hologram, which reflects the reproduction illumination light which has been emitted from a reproduction light source and which has passed through the reflection hologram, which emits the reflected reproduction illumination light toward the reflection hologram from the first side, and which transmits light of a reproduced image reproduced by the reflection hologram.

Abstract: An image recognition device adds a margin region filled with predetermined image data to a periphery of a captured image captured by a camera to create an expanded image larger than the captured image. When a person is too close to the camera, a part of a person image protrudes from the captured image, but a large detection window which also includes the margin region is set, whereby a window image including a protruding region is taken out of the expanded image. The window image lacks a protruding part of the person image, but it is an image showing an entire body of the person image. The image recognition device stores many reference images assuming various states of a person. The image recognition device extracts features from the window image, and compares them with a feature of the reference image, thereby recognizing the person image.

Abstract: [Problem] To provide a moving body for which it is possible to change the movement direction without changing the orientation during travel. [Solution] With a moving body 1, an exterior case 2 and a drive part 3 are supported by a rotating shaft 4 so as to be able to rotate relative to each other. An upper support shaft 8 of the rotating shaft 4 and the exterior case 2 are connected at a position for which the center of gravity of the exterior case 2 is on the axis of a lower support shaft 10 of the rotating shaft 4. The weight of the exterior case 2 is such that a sufficiently high moment of inertia is provided so that no rotation occurs even when the drive part 3 rotates against the rotational resistance of a bearing 9.

Abstract: According to a moving body, position and direction of a user are calculated from distance measurement data, and on the basis thereof, the position and direction of the moving body are calculated. If the moving body is located in the first or fourth quadrant of the user coordinate system, a right-side moving target is set as the moving target. If the moving body is in the second or the third quadrant of the user coordinate system, a left-side moving target is set as the moving target. Accordingly, the moving target is set on the basis of the position of the moving body with respect to the user, so the moving body is prevented from cutting across in front of the user while its travelling.

Abstract: An image recognition device includes: an image processing device that acquires a feature amount from an image; and an identification device that determines whether a prescribed identification object is present in the image, and identifies the identification object. The identification device includes a BNN that has learned the identification object in advance, and performs identification processing by performing a binary calculation with the BNN on the feature amount acquired by the image processing device. Then, the identification device selects a portion effective for identification from among high-dimensional feature amounts output by the image processing device to reduce the dimensions used in identification processing, and copies low-dimensional feature amounts output by the image processing device to increase dimensions.

Abstract: A pulse wave detection device detects the face from a frame image and corrects the brightness of the face in accordance with the size of the face surface to appropriately detect a pulse wave even if brightness is changed due to movement of a subject with respect to lighting. The brightness of the face depends on the distance between the lighting and the face, and this is equal to indirect measurement of the distance from the lighting to the face using the size of the face. As described above, a pulse wave detecting device estimates/measures a pulse rate of the subject from a brightness signal after intensity of the brightness of the face photographed on the frame image is corrected using the size of the face, by using the correlation between the size of the face of the subject on the frame image and the brightness of the face.

Abstract: An image recognition device includes: an image processing device that acquires a feature amount from an image; and an identification device that determines whether a prescribed identification object is present in the image, and identifies the identification object. The identification device includes a BNN that has learned the identification object in advance, and performs identification processing by performing a binary calculation with the BNN on the feature amount acquired by the image processing device. Then, the identification device selects a portion effective for identification from among high-dimensional feature amounts output by the image processing device to reduce the dimensions used in identification processing, and copies low-dimensional feature amounts output by the image processing device to increase dimensions.

Abstract: A vehicle includes: N (N is an integer equal to or larger than 3) wheels including a pair of wheels spaced apart from each other in a width direction of the vehicle and a steered wheel turnable to right and left relative to the vehicle body; operation input unit to be operated to input a steering direction of the steered wheel; lean mechanism for leaning the vehicle body in its width direction; and lean control unit for controlling the lean mechanism. Lean control unit controls the lean mechanism so that vehicle body leans toward the steering direction side in response to an input into the operation input unit. Vehicle includes detector for detecting a force that changes a lean angle of the vehicle body. Lean control unit causes the lean mechanism to generate a torque that produces a force in a direction opposite to a force detected by the detector.

Abstract: According to a moving body, the position and orientation of a user are calculated based on distance measurement data which is the measured distance to the user, and based on this, a moving target and a control target for the moving body with respect to the user are calculated. In addition, the position and the orientation of the moving body in a user coordinate system are calculated based on the position and direction of the user, and a shift control target which is set by shifting the control target toward the user side is calculated based on an angular deviation, which is the angle formed by the orientation of the moving body. A drive unit is operated such that the moving body moves toward the shift control target, so the moving body can be prevented from moving in a large turn when the user turns or rotates.

Abstract: From distance measurement data obtained by a distance measurement, according to a moving body approximate ellipses and quadratic functions along an xy-axis and yx-axis are calculated. An approximate ellipse close to the shoulder width and thickness of a user is selected, and quadratic functions having the smallest approximation error to the distance measurement data are selected. Thus, the position and orientation of the user are estimated from the selected approximate curves. That is, if an approximate curve based on one shape or coordinate system does not match the user, the position and orientation of the user are estimated from an approximate curve based on another shape or coordinate system. Therefore, even if the positional relation between the distance measurement sensor and user changes variously, the position and orientation of the user can be detected.

Abstract: A spatio-temporal image recognition device includes spatio-temporal image data generation units for converting moving-image data which continuously holds spatial information and temporal information to spatio-temporal image data, and they scan the moving-image data on scanning paths different from each other. The spatio-temporal image data generation units generate spatio-temporal image data scanned on the scanning paths different from each other and output them to an image recognition unit. The image recognition unit generates two-dimensional feature maps by individual convolution process of the spatio-temporal image data and then, integrates them, analyzes them by a neural network, and outputs an image recognition result.