Abstract: A map database creation method is provided. The method includes: obtaining a factor set including factors; dividing a map database into levels based on the factors, and taking each interval of the last level as one sub-database; creating an initial map based on a factor value of each factor corresponding to each sub-database, and creating the sub-database as an initial map database by storing the corresponding initial map in the sub-database; finding the initial map matching a current lighting condition from the initial map database based on the current lighting condition, and taking the found initial map as a positioning map; and performing a visual positioning based on the positioning map, creating an expanded map corresponding to the current lighting condition based on the visual positioning, and creating the sub-database corresponding to the current lighting condition as an expanded map database by storing the corresponding expanded map in the sub-database.

Abstract: A backlight face recognition method, a terminal device using the same, and a computer readable storage medium are provided. The method includes: performing a face detection on each original face image in an original face image sample set to obtain a face frame corresponding to the original face image; capturing the corresponding original face images from the original face image sample set, and obtaining a new face image containing background pixels corresponding to the captured original face images from the original face image sample set; preprocessing all the obtained new face images to obtain a backlight sample set and a normal lighting sample set; and training a convolutional neural network using the backlight sample set and the normal lighting sample set until the convolutional neural network reaches a preset stopping condition. The trained convolutional neural network will improve the accuracy of face recognition in complex background and strong light.

Abstract: A robot control method includes: obtaining force information associated with a left foot and a right foot of the robot; calculating a zero moment point of a COM of a body of the robot based on the force information; updating a motion trajectory of the robot according to the zero moment point of the COM of the body to obtain an updated position of the COM of the body; performing inverse kinematics analysis on the updated position of the COM of the body to obtain joint angles of a left leg and a right leg of the robot; and controlling the robot to move according to the joint angles.

Abstract: A computer-implemented gait planning method includes: determining a pitch angle between a foot of the robot and a support surface where the robot stands; determining a support point on a sole of the foot according to the pitch angle; calculating an ankle-foot position vector according to the support point, wherein the ankle-foot position vector is a position vector from an ankle of the robot to a support point on a sole of the foot; calculating a magnitude of change of an ankle position according to the pitch angle and the ankle-foot position vector; and obtaining a compensated ankle position by compensating the ankle position according to the magnitude of change of the ankle position.

Abstract: A mechanical arm includes a first link connectable to a surface, a second link, a third link, a fourth link, and a fifth link that are coupled to one another in series, and an end effector connectable to the fifth link. The end effector is rotatable about an axis of rotation same as an axis of rotation of the fourth link, and rotatable about an axis of rotation orthogonal to the axis of rotation of the fourth link. The first link, the second link, the third link, the fourth link, and the fifth link are collectively structured and configured to rotate such that the end effector is actuatable to a workspace under the surface.

Abstract: A direct force feedback control method as well as a controller and a robot using the same are provided. The method includes: obtaining an actual position and an actual speed of an end of the robotic arm and an actual external force acting on the end in a Cartesian space; calculating an impedance control component of the end in the Cartesian space based on the obtained actual position, the obtained actual speed, the obtained actual external force, an expected position, an expected speed, and an expected acceleration of the end; calculating a force control component of the end in the Cartesian space based on an expected interaction force acting on the end, the actual external force, and the actual speed; determining whether the actual external force is larger than a preset threshold, and obtaining a total force control quantity of the end of the robotic arm in the Cartesian space.

Abstract: The present disclosure provides a robot mapping method as well as a robot and a computer readable storage medium using the same. The method includes: detecting a marker with identification information capable of being identified by the robot in a current scene; determining whether the detected marker meets a preset condition; and mapping the current scene based on the marker, if the detected marker meets the preset condition. The robot mapping method can not only map the current scene, but also effectively reduce the difficulty of loops and the number of false loops.

Abstract: The present disclosure provides a redundant robotic arm control method, a redundant robotic arm, and a computer readable storage medium. The method includes: obtaining an external force acting on an end of the robotic arm and an external torque acting on each joint; calculating a first joint speed of each joint based on a degree of influence of the joint on the end in each motion dimension and the external force acting on the end; determining a zero space speed of each joint corresponding to a current position of the end based on a link torque of an external force acting on a link with respect to the joint; calculating a total joint speed based on the first joint speed and the zero space speed; and controlling the robotic arm to the move according to the total joint speed.

Abstract: A robot climbing control method is disclosed. A gravity direction vector in a gravity direction in a camera coordinate system of a robot is obtained. A stair edge of stairs in a scene image is obtained and an edge direction vector of the stair edge in the camera coordinate system is determined. A position parameter of the robot relative to the stairs is determined according to the gravity direction vector and the edge direction vector. Poses of the robot are adjusted according to the position parameter to control the robot to climb the stairs.

Abstract: The present disclosure provides a method for controlling end-portions of a robot. The method includes obtaining joint information of a robot by at least one sensor and determining a first posture of an end-portion of the robot in accordance with the joint information, obtaining end-portion information of the robot by the sensor and obtaining the second posture of the end-portion of the robot including the interference information in accordance with the end-portion information of the robot and the first posture of the end-portion of the robot, and conducting a closed-loop control on the robot in accordance with an error between the second posture of the end-portion of the robot and a predetermined expected posture of the end-portion of the robot.

Abstract: A humanoid robot and its balance control method and computer readable storage medium are provided. Expected accelerations of each of a sole and centroid of a humanoid robot corresponding to a current expected balance trajectory and an expected angular acceleration of the waist corresponding to the current expected balance trajectory are obtained based on current motion data of the sole, the centroid, and the waist, respectively first, then an expected angular acceleration of each joint meeting control requirements of the sole, the centroid, and the waist while the robot corresponds to the current expected balance trajectory is calculated based on an angular velocity of the joint, the expected accelerations of the waist, the sole, and the centroid, respectively, and then each joint of the robot is controlled to move at the obtained expected angular acceleration of the joint based on the angular displacement of the joint.

Abstract: A computer-implemented method for text conversion, a computer device, and a non-transitory computer readable storage medium are provided. The method includes: obtaining a text to be converted; performing a non-standard word recognition on the text to be converted, to determine whether the text to be converted includes a non-standard word; recognizing the non-standard word in the text to be converted by using an eXtreme Gradient Boosting model in response to the text to be converted including the non-standard word; and obtaining a target converted text corresponding to the text to be converted, according to a recognition result outputted by the eXtreme Gradient Boosting model. The method has a faster recognition speed and a higher recognition accuracy compared with the deep learning model.

Abstract: A robotic finger structure includes a proximal phalanx; a middle phalanx rotatably connected to one end of the proximal phalanx; a distal phalanx rotatably connected to one end of the middle phalanx and defining a distal phalanx opening in a front side thereof and at one end adjacent to the middle phalanx; a connecting rod having opposite ends that are rotatably connected to the proximal phalanx and the distal phalanx, and an actuating assembly to drive the middle phalanx to rotate with respect to the proximal phalanx. The connecting rod includes a first angled segment having a first recess facing a back side of the middle phalanx. When the distal phalanx is flush with the middle phalanx, the first angled segment passes through the distal phalanx opening, and a first end of the distal phalanx opening extends into the first recess.

Abstract: A dynamic gesture recognition method includes: performing detection on each frame of image of a video stream using a preset static gesture detection model to obtain a static gesture in each frame of image of the video stream; in response to detection of a change of the static gesture from a preset first gesture to a second gesture, suspending the static gesture detection model and activating a preset dynamic gesture detection model; and performing detection on multiple frames of images that are pre-stored in a storage medium using the dynamic gesture detection model to obtain a dynamic gesture recognition result.

Abstract: The present disclosure is provides a method, an apparatus and a terminal device for constructing parts together. The method includes: determining an assembly progress of an object constructed of parts; determining a currently required part according to the assembly progress; identifying, in a part photo of the object, the currently required part using a pre-trained first neural network model and marking the one or more identified currently required parts in the part photo; and displaying, via a display, a 3D demonstration animation with the marked part photo as a background image. The method is based on the existing 3D demonstration animation, which identifies the currently required part in the captured part photo and marks the identified currently required part in the part photo, and then displays the marked part photo as the background image of the 3D demonstration animation, thereby adding a prompt for the real part.

Abstract: The present disclosure discloses a navigation map updating method as well as an apparatus, and a robot using the same. The method includes: controlling a robot to move along a designated path after a successful relocalization of the robot, and recording key frame data of each frame on the designated path and a corresponding pose; creating a new navigation map, and copying information in an original navigation map into the new navigation map; and covering the key frame data of each frame on the designated path onto the new navigation map to obtain an updated navigation map. In this manner, there is no need for the user to manipulate the robot to recreate the map at the environment where the robot is operated, which saves a lot of time and manpower.