2138 results about "Pose" patented technology

Three-dimensional position information is used to identify the gesture created by a body part of interest. At one or more instances of an interval, the posture of a body part is recognized, based on the shape of the body part and its position and orientation. The posture of the body part over each of the one or more instances in the interval are recognized as a combined gesture. The gesture is classified for determining an input into a related electronic device.

The present invention provides a system and method for recognizing a 3D object in a single camera image and for determining the 3D pose of the object with respect to the camera coordinate system. In one typical application, the 3D pose is used to make a robot pick up the object. A view-based approach is presented that does not show the drawbacks of previous methods because it is robust to image noise, object occlusions, clutter, and contrast changes. Furthermore, the 3D pose is determined with a high accuracy. Finally, the presented method allows the recognition of the 3D object as well as the determination of its 3D pose in a very short computation time, making it also suitable for real-time applications. These improvements are achieved by the methods disclosed herein.

The invention is related to methods and apparatus that use a visual sensor and dead reckoning sensors to process Simultaneous Localization and Mapping (SLAM). These techniques can be used in robot navigation. Advantageously, such visual techniques can be used to autonomously generate and update a map. Unlike with laser rangefinders, the visual techniques are economically practical in a wide range of applications and can be used in relatively dynamic environments such as environments in which people move. One embodiment further advantageously uses multiple particles to maintain multiple hypotheses with respect to localization and mapping. Further advantageously, one embodiment maintains the particles in a relatively computationally-efficient manner, thereby permitting the SLAM processes to be performed in software using relatively inexpensive microprocessor-based computer systems.

The invention relates to an autonomously identifying and capturing method of a non-cooperative target of a space robot, comprising the main steps of (1) pose measurement based on stereoscopic vision, (2) autonomous path planning of the target capture of the space robot and (3) coordinative control of a space robot system, and the like. The pose measurement based on the stereoscopic vision is realized by processing images of a left camera and a right camera in real time, and computing the pose of a non-cooperative target star relative to a base and a tail end, wherein the processing comprises smoothing filtering, edge detection, linear extraction, and the like. The autonomous path planning of the target capture of the space robot comprises is realized by planning the motion tracks of joints in real time according to the pose measurement results. The coordinative control of the space robot system is realized by coordinately controlling mechanical arms and the base to realize the optimal control property of the whole system. In the autonomously identifying and capturing method, a self part of a spacecraft is directly used as an identifying and capturing object without installing a marker or a comer reflector on the target star or knowing the geometric dimension of the object, and the planned path can effectively avoid the singular point of dynamics and kinematics.

Providing information associated with a view of a real environment superimposed with a virtual object, includes obtaining a model of a real object located in a real environment and a model of a virtual object, determining a spatial relationship between the real object and the virtual object, determining a pose of a device displaying the view relative to the real object, determining a visibility of the virtual object in the view according to the pose of the device, the spatial relationship and the models of the real object and the virtual object, and if the virtual object is not visible in the view, determining a movement for at least one of the device and at least part of the real object such that the virtual object is visible in the view in response to the at least one movement, and providing information indicative of the at least one movement.

The invention discloses an aircraft component position and pose adjusting system based on four locators and a method thereof. The position and pose adjusting system comprises four three-coordinate locators, a spherical technical connector, an aircraft component to be adjusted, a laser tracker and a target reflecting sphere, the three-coordinate locator comprises a bottom plate, and an X-direction motion mechanism, a Y-direction motion mechanism, a Z-direction motion mechanism and a displacement sensor which are arranged from the lower part and the upper part in sequence. The position and pose adjusting method comprises the following steps: firstly, a global coordinate system OXYZ is established, and the current position and pose and the target position and pose of the aircraft component to be adjusted are calculated; secondly, the path of the aircraft component to be adjusted from the current position and pose to the target position and pose is planed; thirdly, the tracks of motion mechanisms in all the directions are formed according to the path; and fourthly, the three locators are synergetically moved, and the position adjusting is realized. The method has the following advantages: firstly, the supporting to the aircraft component to be adjusted can be realized; secondly, the automatic adjusting to position and pose of the aircraft component to be adjusted can be realized; and thirdly, the inch adjusting of position and pose of the aircraft component to be adjusted can be realized.

The invention relates to a cascaded convolutional neural network-based quick detection method of an irregular-object grasping pose of a robot. Firstly, a cascaded-type two-stage convolutional-neural-network model of a position-attitude rough-to-fine form is constructed, in a first stage, a region-based fully convolutional network (R-FCN) is adopted to realize grasping positioning and rough estimation of a grasping angle, and in a second stage, accurate calculation of the grasping angle is realized through constructing a new Angle-Net model; and then current scene images containing to-be-grasped objects are collected to be used as original on-site image samples to be used for training, the two-stage convolutional-neural-network model is trained by means of a transfer learning mechanism, then each collected monocular color image is input to the cascaded-type two-stage convolutional-neural-network model in online running, and finally, an end executor of the robot is driven by an obtainedgrasping position and attitude for object grasping control. According to the method, grasping detection accuracy is high, detection speed of the irregular-object grasping pose of the robot is effectively increased, and real-time performance of running of a grasping attitude detection algorithm is improved.

An apparatus is provided with a first sensor unit that obtains two-dimensional information or three-dimensional information about a target object with a first position and orientation, a second sensor unit that obtains the two-dimensional information about the target object, a three-dimensional position and orientation measurement unit that measures three-dimensional position and orientation of the target object based on the information obtained by the first sensor unit, a second sensor position and orientation determination unit that calculates second position and orientation based on a measurement result with the three-dimensional position and orientation measurement unit and model information about the target object, and a three-dimensional position and orientation measurement unit that measures the three-dimensional position and orientation of the target object based on the information obtained by the second sensor unit with the second position and orientation and the model information about the target object.

The present invention provides a multi-camera-based collaborative visual SLAM method, in particular to a collaborative visual SLAM method using multiple cameras in a dynamic environment. The method of the invention allows the relative position and direction of the cameras to change with time, so that multiple cameras can move independently and be installed on different platforms, and solve problems related to camera pose estimation, map point classification and camera group management, etc. problem, so that the method works robustly in dynamic scenes and can reconstruct the 3D trajectories of moving objects. Compared with the existing SLAM method based on single camera, the method of the present invention is more accurate and robust, and can be applied to micro-robots and wearable augmented reality.

The invention relates to an online offset correction method and device for robot hand-eye calibration. The online offset correction method comprises the steps that coordinate values of the circle centers of nine circles on a calibration plate in a camera coordinate system and a base coordinate system are obtained; a transformation equation of the circle center of each circle on the calibration plate from the camera coordinate system to the robot base coordinate system is established, through offset coordinates of each circle, the least square method is adopted to calculate a homogeneous transformation matrix of the camera coordinate system relative to the robot base coordinate system; and according to calibrated pose values of the camera coordinate system relative to the base coordinate system, errors of the offset correction calibration results of the nine circles on the calibration plate are analyzed through a vector two-norm formula, and thus the precision of the online offset correction method is evaluated. According to the online offset correction method and device for robot hand-eye calibration, offset in the robot hand-eye calibration process is corrected, flexible, preciseand rapid adjustment on a production line can be achieved, high-repeatability and precise grabbing operation can be achieved, the online offset correction method and device can be applied to operationof an SCARA robot hand-eye device, and the simple, efficient and high-precision effects are achieved.

It is the measuring apparatus and method for the space position of rigid body and belongs to the field of observation and control. The invention provides a contact measuring apparatus for space position of a rigid body of low cost and high accuracy, which comprises measuring execution structure, data-collecting apparatus and a computer capable of storing calculation programs. The invention also discloses a method for the measuring. The method is used to get space position information of the rigid body to be measured by the use of iterative equations set according to the initial datum obtained from the said apparatus. The said apparatus is characterized by simple structure and high system real-time capabilities and can realize the closed-loop real-time control of the rigid body to be measured.

The invention discloses a robot autonomous classification grabbing method based on YOLOv3, and the method is characterized in that the method comprises: collecting and constructing a target object sample data set; training the YOLOv3 target detection network to obtain a target object recognition model; acquiring a color image and a depth image of the target object; processing the color image by using the trained YOLOv3 target detection network to obtain the category information and the position information of the target object to be grabbed, and further processing by combining the depth imageto obtain the point cloud information of the target object; and performing minimum bounding box solving on the point cloud information, calculating the main direction of the point cloud by combining aPCA algorithm, calibrating the coordinate data of the X, Y and Z axes of the target object, and calculating the six-degree-of-freedom pose of the target object relative to the robot coordinate system. According to the method, the YOLOv3 algorithm is adopted, the object grabbing pose is estimated through point cloud preprocessing, PCA and other methods, and then the robot grabs the target objectsin a classified mode.

The invention provides a visual 3D taking and placing method and system based on a cooperative robot. The method comprises the steps that internal and external parameters of a camera of a binocular structured light three-dimensional scanner are calibrated; the hands and eyes of the cooperative robot are calibrated, and a calibration result matrix is obtained; a three-dimensional digital model of to-be-taken-and-placed target objects is collected; the calibrated binocular structured light three-dimensional scanner is adopted to obtain point cloud data of the to-be-taken-and-placed target objects which are stacked in a scattered mode, and the point cloud is segmented to obtain scene point clouds of the multiple to-be-taken-and-placed target objects; the to-be-taken-and-placed target object with the highest grabbing success rate is selected as a grabbing target according to the scene point clouds of the multiple to-be-taken-and-placed target objects; the three-dimensional digital model ofthe grabbing target and scene point pair features are registered, pre-defined taking and placing pose points are registered into a scene, and a registered pose estimation result is obtained and serves as a grabbing pose of the grabbing target; and a preliminary grabbing path track of the cooperative robot is planned. The target object can be accurately recognized, and the grabbing positioning precision is high.

The invention relates to the field of cross fusion of computer vision and deep learning, in particular to a semantic mapping method based on visual SLAM and two-dimensional semantic segmentation. Themethod comprises the following steps: S1, calibrating camera parameters, and correcting camera distortion; S2, acquiring an image frame sequence; S3, preprocessing the image; S4, judging whether the current image frame is a key frame or not, if so, turning to the step S6, and if not, turning to the step S5; S5, performing dynamic fuzzy compensation; s6, carrying out semantic segmentation, extracting ORB feature points for the image frames, and carrying out semantic segmentation by using a mask region convolutional neural network algorithm model; s7, pose calculation: utilizing a sparse SLAM algorithm model to calculate the pose of the camera; s8, using the semantic information for assisting in dense semantic map construction, and achieving three-dimensional semantic map construction of theglobal point cloud map. According to the invention, the performance of the unmanned aerial vehicle semantic mapping system can be improved, and the robustness of feature point extraction and matchingfor a dynamic scene is significantly improved.

The invention relates to the technical field of mode recognizing, in particular to a real-time human body action recognizing method and device based on depth image sequence. The method comprises the steps of S1, extracting target action sketch from a target depth image sequence and extracting a training action sketch from a training depth image set; S2, performing gesture clustering on the training action sketch and performing action calibrating on the clustered outcome; S3, computing the gesture characteristics of the target action sketch and training action sketch; S4, performing the gesture training based on a Gauss mixing model by combining the gesture characteristics of the training action sketch and constructing a gesture model; S5, computing the transferring probability among all gestures of the clustered outcome in each action and constructing an action image model; and S6, performing action recognizing on the target depth image sequence according to the gesture characteristics of the target action sketch, the gesture model and the action image model. The real-time human body action recognizing method disclosed by the invention has the advantages of improving the efficiency of action recognizing and the accuracy and the robustness of the action recognizing.

Robotic mobility assistant exoskeleton with frame members, which are attached adjacent to biological joints, supplements relative movement between skeletal members. Mechanical joint defines a center of relative rotation of frame members about three mutually perpendicular axes with the center of relative rotation of frame members being displaced from the outer surfaces of skeletal members to correspond in position with the center of biological joint. Actuation devices including powered cable springs rotate frame members. Control system executes calibrated user specific posture sequences and activates power. Mechanical joint exhibits cylindrical guide surfaces defining a slidable or rotatable connection of frame members and has a radius intersecting the center of the biological joint. Node control network distributes computing load and reduces communications overhead. Map unit has object recognition system for monitoring environment. Standard posture data derived from rules of ambulation and posture provide positioning control, optimized by energy-timed margins for balance maintenance.

Systems, methods and products for animating non-humanoid characters with human motion are described. One aspect includes selecting key poses included in initial motion data at a computing system; obtaining non-humanoid character key poses which provide a one to one correspondence to selected key poses in said initial motion data; and statically mapping poses of said initial motion data to non-humanoid character poses using a model built based on said one to one correspondence from said key poses of said initial motion data to said non-humanoid character key poses. Other embodiments are described.

The invention discloses a novel semantic mapping system based on instant positioning and mapping and three-dimensional point cloud semantic segmentation, and belongs to the technical field of computervision and artificial intelligence. According to the method, a sparse map is established by utilizing instant positioning and mapping, key frames and camera poses are obtained, and semantic segmentation is carried out based on the key frames by utilizing point cloud semantic segmentation. A two-dimensional target detection method and point cloud splicing are utilized to obtain a truncated body suggestion, a Bayesian updating scheme is designed to integrate semantic tags of candidate truncated bodies, and points with final correction tags are inserted into an established sparse map. Experiments show that the system has high efficiency and accuracy.