A robot control method includes: determining a planned capture point and a measured capture point of the robot so as to calculate a capture point error of the robot; obtaining positions of a left foot and a right foot of the robot, and a planned zero moment point (ZMP) of the robot so as to calculate desired support forces of the left foot and the right foot; calculating desired torques of the left foot and the right foot according to the capture point error, the desired support forces of the left foot and the right foot; obtaining measured torques of the left foot and the right foot so as to calculate desired poses of the left foot and the right foot; and controlling the robot to walk according to the desired poses of the left foot and the desired pose of the right foot.

A method for controlling a robotic device includes observing a first object associated with an object type at one or more first locations in an environment over a period of time prior to a current time. The method also includes generating a probability distribution associated with the one or more first locations based on observing the first object over the period of time. The method further includes observing, at the current time, a second object associated with the object type at a second location in the environment. The method still further includes determining a probability of the second object being at the second location based on observing the second object at the second location. The probability is based on the probability distribution associated with the one or more first locations. The method also includes controlling the robotic device to perform an action based on the probability being less than a threshold.

A biped robot gait control method as well as a robot and a computer readable storage medium are provided. During the movement, the system obtains a current supporting pose of a current supporting leg of the biped robot, and calculates a relative pose between the supporting legs based on the current supporting pose and a preset ideal supporting pose of a next step. The system further calculates modified gait parameters of the next step based on the relative pose between the two supporting legs and a joint distance between left and right ankle joints in an initial state of the biped robot when standing. Finally, the system controls the next supporting leg to move according to the modified gait parameters.

A pose determination method and a robot using the same are provided. The method includes: obtaining a two-dimensional code image collected by the camera of the robot and sensor data collected by the sensor of the robot, and determining mileage information of the robot within a predetermined duration, where the sensor data includes an acceleration and an angular velocity, determining a first pose of the camera based on two-dimensional code information recognized from the two-dimensional code image and a pose estimation function, and determining a second pose of the sensor based on the sensor data; obtaining a third pose by performing a tight coupling optimization based on the first pose and the second pose; and obtaining the pose of the robot by fusing the third pose and the mileage information. In such a manner, the accuracy of determining the pose of the robot in a complex scene can be improved.

The present disclosure discloses a robot control method as well as an apparatus, and a robot using the same. The method includes: obtaining a human pose image; obtaining pixel information of key points in the human pose image; obtaining three-dimensional positional information of key points of a human arm according to the pixel information of the preset key points; obtaining a robotic arm kinematics model of a robot; obtaining an angle of each joint in the robotic arm kinematics model according to the three-dimensional positional information of the key points of the human arm and the robotic arm kinematics model; and controlling an arm of the robot to perform a corresponding action according to the angle of each joint. The control method does not require a three-dimensional stereo camera to collect three-dimensional coordinates of a human body, which reduces the cost to a certain extent.

A system providing dynamic balancing in a robotic system. The system includes memory storing a definition of a robot and storing an input animation for the robot specifying motion of components of the robot. A simulator performs a dynamic simulation of the robot performing the input animation including modeling a first set of the components as flexible components and a second set of the components as rigid components. Each of the flexible components is coupled at opposite ends to one of the rigid components. An optimizer generates a retargeted motion for the components to provide dynamic balancing of the robot performing the retargeted motion. The optimizer generates the retargeted motion by transforming forces acting on the robot to a local contact frame rigidly moving with the robot. The optimizer generates the retargeted motion so a zero-moment point of the robot lies in a support area of the robot's feet.

A feedforward control method comprising steps of: acquiring kinematic parameters of each joint of a robot based on inverse kinematics according to a pre-planned robot motion trajectory, and setting a center of a body of the robot as a floating base; determining a six-dimensional acceleration of a center of mass of each joint of the robot in a base coordinate system using a forward kinematics algorithm, based on the kinematic parameters of each joint of the robot, and converting the six-dimensional acceleration of the center of mass of each joint of the robot in the base coordinate system to a six-dimensional acceleration in a world coordinate system; and calculating a torque required by a motor of each joint of the robot using an inverse dynamic algorithm, and controlling the motors of corresponding joints of the robot.

A large area surveillance method is based on weighted double deep Q-learning. A robot which of Q-value table including a Q.sub.A-value table and Q.sub.B-value table is provided, an unidentified object enters a large space to trigger the robot, and the robot perceives a current state s and determines whether the current state s is a target state, if yes, the robot reaches a next state and monitors the unidentified object, and if not, the robot reaches a next state, obtains a reward value according to the next state, selectively updates a Q.sub.A-value or Q.sub.B-value with equal probability, and then updates a Q-value until convergence to obtain an optimal surveillance strategy. The problems of a limited surveillance area and camera capacity are resolved, and the synchronization of multiple cameras doesn't need to be considered, and thus the cost is reduced. A large area surveillance robot is also disclosed.

The present disclosure provides a robot posture control method as well as a robot and a computer readable storage medium using the same. The method includes: constructing a virtual model of the robot, wherein the virtual model comprises a momentum wheel inverted pendulum model of the robot and an angle between a sole surface of the robot and a horizontal plane; and performing a posture control based on outer-loop feedback control, inner loop compensation for the external disturbance rejection in position level, inner loop external disturbance rejection via null-space in velocity level, and inner loop external disturbance rejection in force/acceleration level on the robot. In this manner, a brand-new virtual model is provided, which can fully reflect the upper body posture, centroid, foot posture, and the like of the robot which are extremely critical elements for the balance and posture control of the robot.

Systems and methods for online augmentation for learned grasping are provided. In one embodiment, a method is provided that includes identifying an action from a discrete action space. The method includes identifying a second set of grasps of the agent utilizing a transition model based on the action and at least one contact parameter. The at least one contact parameter defines allowed states of contact for the agent. The method includes applying a reward function to evaluate each grasp of the second set of grasps based on a set of contact forces within a friction cone that minimizes a difference between an actual net wrench on the object and a predetermined net wrench. The reward function is optimized online using a lookahead tree. The method includes selecting a next grasp from the second set. The method includes causing the agent to execute the next grasp.