Abstract: A foothold position control system and method for a biped robot are provided. 1) A feasible collision-free path is planned by using a path planning algorithm; 2) an available foothold area of a swing foot is determined according to step-length constraints, movement capabilities, foot sizes, and center offsets of a biped robot; and 3) fuzzy processing is performed to determine a specific foothold position of the biped robot. Selection of suitable foothold positions on both sides of a path when a biped robot executes specific walking actions after finishing path planning is realized. The foothold position control system and method has the advantages of being simple and easy to implement, having low computational load and high speed, being capable of exerting extreme movement capabilities of different biped robots, enabling more flexible movement of the biped robots, and so on.

Abstract: A robot joint torque control system and a load compensation method therefor are provided, which relate to the technical field of robot joint motion control. A mathematical model of the robot joint torque control system is established first. Equivalent transformation is performed on a system functional block diagram thereof, and then it can be seen that load parameters have a great influence on joint torque output. A load compensation controller is designed to effectively eliminate the influence of the load parameters on an output torque of the joint. The system is equivalent to an inertial element on the basis of the compensation, and then a PD controller parameter is adjusted to increase an open-loop gain of the system, so as to increase a system bandwidth and increase a response speed of the joint torque control system, thereby improving performance of the joint torque control system.

Abstract: A foothold position control system and method for a biped robot are provided. 1) A feasible collision-free path is planned by using a path planning algorithm; 2) an available foothold area of a swing foot is determined according to step-length constraints, movement capabilities, foot sizes, and center offsets of a biped robot; and 3) fuzzy processing is performed to determine a specific foothold position of the biped robot. Selection of suitable foothold positions on both sides of a path when a biped robot executes specific walking actions after finishing path planning is realized. The foothold position control system and method has the advantages of being simple and easy to implement, having low computational load and high speed, being capable of exerting extreme movement capabilities of different biped robots, enabling more flexible movement of the biped robots, and so on.

Abstract: The present invention discloses a method of tracking control for a foot force and moment of a biped robot. According to the method, a double-spring damping model is designed, and a force tracking controller is designed by using an LQR optimization method, so as to realize tracking of the foot force and moment of the biped robot. Further, a desired force on a foot and a desired moment on the foot are calculated through a planned ZMP distribution method, thereby eventually achieving better ZMP tracking of the biped robot and adapting to ground of certain unevenness.

Abstract: The present application discloses a wide-field-of-view anti-shake high-dynamic bionic eye. A trajectory tracking method based on a bionic eye robot includes: establishing a linear model according to a bionic eye robot; establishing a full state feedback control system on the basis of the linear model; in the full state feedback control system, acquiring an angle and an angular acceleration required for a joint in a target tracking process of the bionic eye on the basis of a preset trajectory expectation value and a preset joint angle expectation value; the method further includes: adopting a linear quadratic regulator (LQR) to calculate a parameter K in the full state feedback control system, and minimizing energy consumption by establishing an energy function, so as to optimize the coordinated head-eye motion control of the linear bionic eye. The present application achieves the optimal control of the target tracking.

Abstract: The present invention provides a method for monitoring a balanced state of a humanoid robot, comprising: acquiring state data of the robot falling in different directions and being stable, forming a support vector machine (SVM) training data set and obtaining, by training, an initial SVM classifier; inputting the state data of the robot to the trained SVM classifier, so that the SVM classifier outputs a classification result; taking statistics on a proportion of cycles judged to have an impending fall in the total number of control cycles within a judgment buffer time after the SVM classifier outputs the classification result, and finally determining a monitoring result of the balanced state of the robot according to the proportion and finally extracting state data of misjudged cycles within the buffer time, adding the state data to the current training data set and updating the SVM classifier, eventually enabling the classifier to achieve the effects of matching motion capabilities of the robot and monitorin

Abstract: The present application discloses a wide-field-of-view anti-shake high-dynamic bionic eye. A trajectory tracking method based on a bionic eye robot includes: establishing a linear model according to a bionic eye robot; establishing a full state feedback control system on the basis of the linear model; in the full state feedback control system, acquiring an angle and an angular acceleration required for a joint in a target tracking process of the bionic eye on the basis of a preset trajectory expectation value and a preset joint angle expectation value; the method further includes: adopting a linear quadratic regulator (LQR) to calculate a parameter K in the full state feedback control system, and minimizing energy consumption by establishing an energy function, so as to optimize the coordinated head-eye motion control of the linear bionic eye. The present application achieves the optimal control of the target tracking.

Abstract: The present disclosure provides a motor stator and a motor. The motor stator comprises a core; windings wound on the core; and a three-dimensional fitted winding inner groove embedding structure which is a cage-type structure having ribs and ends, the ribs being embedded in grooves between the windings, and the ends being connected to the ribs and cover ends of stator windings; both of the rib and the end of the inner groove embedding structure comprise nonmagnetic heat-conducting glue layers and a phase-change material layer interposed as an interlayer between the heat-conducting glue layers. According to the present disclosure, the heat of each part of the motor stator can be quickly conducted, so as to rapidly reduce the temperature, quickly eliminate the instantaneous temperature peak, and improve the temperature gradient equilibrium of each part.

Abstract: The present disclosure provides a waist joint of humanoid robot and a humanoid robot. The waist joint comprise: tile-shaped structural members, connection members, and substrates. The first end of the first connection member is slidably connected with the first end of the first tile-shaped structural member, and the first end of the second connection member is slidably connected with the second end of the first tile-shaped structural member; one side surface of the first substrate is slidably connected with the first end of the second tile-shaped structural member, and one side surface of the second substrate is slidably connected with the second end of the second tile-shaped structural member; the second end of the first connection member is connected with the third end of the second tile-shaped structural member, and the second end of the second connection member is connected with the fourth end of the second tile-shaped structural member.

Abstract: The present application provides a method and apparatus for pose planar constraining on the basis of planar feature extraction, wherein the method includes: inputting the acquired RGB color image and point cloud image into spatial transformation network to obtain two-dimensional and three-dimensional affine transformation matrixes; extracting the planar features of the transformed two-dimensional affine transformation matrix and three-dimensional affine transformation matrix; inputting the acquired planar features into the decoder and obtain the pixel classification of the planar features; clustering the vectors corresponding to the planar pixels to obtain the segmentation result of the planar sample; using planar fitted by the segmentation result to make planar constraint to the pose calculated by vision algorithm.

Abstract: The present disclosure provides a motor stator and a motor. The motor stator comprises a core; windings wound on the core; and a three-dimensional fitted winding inner groove embedding structure which is a cage-type structure having ribs and ends, the ribs being embedded in grooves between the windings, and the ends being connected to the ribs and cover ends of stator windings; both of the rib and the end of the inner groove embedding structure comprise nonmagnetic heat-conducting glue layers and a phase-change material layer interposed as an interlayer between the heat-conducting glue layers. According to the present disclosure, the heat of each part of the motor stator can be quickly conducted, so as to rapidly reduce the temperature, quickly eliminate the instantaneous temperature peak, and improve the temperature gradient equilibrium of each part.

Abstract: The present invention provides a method for monitoring a balanced state of a humanoid robot, comprising: acquiring state data of the robot falling in different directions and being stable, forming a support vector machine (SVM) training data set and obtaining, by training, an initial SVM classifier; inputting the state data of the robot to the trained SVM classifier, so that the SVM classifier outputs a classification result; taking statistics on a proportion of cycles judged to have an impending fall in the total number of control cycles within a judgment buffer time after the SVM classifier outputs the classification result, and finally determining a monitoring result of the balanced state of the robot according to the proportion and finally extracting state data of misjudged cycles within the buffer time, adding the state data to the current training data set and updating the SVM classifier, eventually enabling the classifier to achieve the effects of matching motion capabilities of the robot and monitorin

Abstract: The present application provides a method and apparatus for pose planar constraining on the basis of planar feature extraction, wherein the method includes: inputting the acquired RGB color image and point cloud image into spatial transformation network to obtain two-dimensional and three-dimensional affine transformation matrixes; extracting the planar features of the transformed two-dimensional affine transformation matrix and three-dimensional affine transformation matrix; inputting the acquired planar features into the decoder and obtain the pixel classification of the planar features; clustering the vectors corresponding to the planar pixels to obtain the segmentation result of the planar sample; using planar fitted by the segmentation result to make planar constraint to the pose calculated by vision algorithm.

Abstract: The application discloses a multi-task learning incorporating dependencies method for bionic eye's face attribute recognition, which is as follows: Determine the first face attribute and the second face attribute for attribute recognition of facial image. Obtain the first recognition task branch and the second recognition task branch. Establish the task dependency between the first recognition task branch and the second recognition task branch to obtain the first transformed face attribute fully connected layer related to the second face attribute. and the second transformed face attribute fully connected layer related to the first face attribute. Feed the first transformed face attribute fully connected layer into the prediction layer to predict the first face attribute of facial image. And feed the second transformed face attribute fully connected layer into the prediction layer to predict the second face attribute of facial image.

Abstract: The present invention discloses a method of tracking control for a foot force and moment of a biped robot. According to the method, a double-spring damping model is designed, and a force tracking controller is designed by using an LQR optimization method, so as to realize tracking of the foot force and moment of the biped robot. Further, a desired force on a foot and a desired moment on the foot are calculated through a planned ZMP distribution method, thereby eventually achieving better ZMP tracking of the biped robot and adapting to ground of certain unevenness.

Abstract: The present disclosure provides a waist joint of humanoid robot and a humanoid robot. The waist joint comprise: tile-shaped structural members, connection members, and substrates. The first end of the first connection member is slidably connected with the first end of the first tile-shaped structural member, and the first end of the second connection member is slidably connected with the second end of the first tile-shaped structural member; one side surface of the first substrate is slidably connected with the first end of the second tile-shaped structural member, and one side surface of the second substrate is slidably connected with the second end of the second tile-shaped structural member; the second end of the first connection member is connected with the third end of the second tile-shaped structural member, and the second end of the second connection member is connected with the fourth end of the second tile-shaped structural member.

Abstract: The application discloses a multi-task learning incorporating dependencies method for bionic eye's face attribute recognition, which is as follows: Determine the first face attribute and the second face attribute for attribute recognition of facial image. Obtain the first recognition task branch and the second recognition task branch. Establish the task dependency between the first recognition task branch and the second recognition task branch to obtain the first transformed face attribute fully connected layer related to the second face attribute. and the second transformed face attribute fully connected layer related to the first face attribute. Feed the first transformed face attribute fully connected layer into the prediction layer to predict the first face attribute of facial image. And feed the second transformed face attribute fully connected layer into the prediction layer to predict the second face attribute of facial image.