Robot control method and apparatus, robot, and storage medium

1 . A robot control method, performed by a robot, and the method comprising:

controlling an end effector of the robot to collide with a target object in a process of the end effector moving to the target object;

controlling the end effector to rotate relative to the target object, to adjust the target object to a target pose corresponding to a pick-up action;

controlling the end effector to pick up the target object after the target object is adjusted to the target pose; wherein the end effector is in motion during an entire phase from the end effector colliding with the target object to picking up the target object;

obtaining, based on a unified equation of state corresponding to a hybrid system comprising the end effector and the target object, pose information and force information of the hybrid system at each time step of the entire phase; and

controlling the end effector at each time step of the entire phase according to the pose information and the force information of the hybrid system at each time step of the entire phase.

2 . The method according to claim 1 , wherein a position where the end effector collides with the target object is located at a bottom of the target object, to allow the target object to fall toward the end effector.

3 . The method according to claim 2 , wherein the controlling the end effector to rotate relative to the target object comprises:

controlling the end effector to rotate in a direction opposite to a falling direction in a process of the target object falling over.

4 . The method according to claim 1 , wherein a pick-up mode of the end effector is non-grasping type in which the target object is held by a palm part of the end effector, and a static friction force on a bearing surface of the end effector allows the target object and the end effector to remain relatively stationary.

5 . The method according to claim 4 , wherein the controlling the end effector to pick up the target object comprises:

controlling the end effector to rotate until the target object and the palm part of the end effector is relatively stationary.

6 . The method according to claim 1 , wherein a pick-up mode of the end effector is grasping type in which a finger part of the end effector is closed toward a palm part of the end effector, to grasp the target object between the palm part and the finger part.

7 . The method according to claim 1 , wherein the pose information and the force information of the hybrid system at each time step of the entire phase meet the following constraint conditions:

an inequality constraint condition, configured to constrain a friction force of a contact position between the end effector and the target object; and

an equality linkage constraint condition, configured to constrain a motion trajectory and an acting force between the end effector and the target object.

8 . The method according to claim 7 , wherein the obtaining, based on the unified equation of state corresponding to the hybrid system comprising the end effector and the target object, the pose information and the force information of the hybrid system at each time step of the entire phase comprises:

constructing an objective function based on a Bellman optimal equation, the unified equation of state, and the constraint conditions; and

solving the objective function by using a primal dual augmented Lagrangian multiplier method, to obtain the pose information and the force information of the hybrid system at each time step of the entire phase.

9 . The method according to claim 1 , wherein the controlling the end effector of the robot to collide with the target object comprises:

obtaining a change amount of a contact velocity between the end effector and the target object before and after a collision of the end effector and the target object, wherein the contact velocity refers to a relative velocity of a contact point between the end effector and the target object;

determining a velocity of the target object after the collision according to the change amount and a velocity of the target object before the collision;

determining a velocity that the end effector needs to reach before the collision, according to the velocity of the target object after the collision, a maximum acceleration of the end effector, and a constraint relationship between the velocity of the target object after the collision and velocities of the end effector before and after the collision; and

controlling the end effector to collide with the target object according to the velocity that the end effector needs to reach before the collision.

10 . A robot control apparatus, comprising: a processor and a memory, the memory storing a computer program, and the computer program being loaded and executed by the processor to implement:

controlling an end effector of the robot to collide with a target object in a process of the end effector moving to the target object;

controlling the end effector to rotate relative to the target object, to adjust the target object to a target pose corresponding to a pick-up action;

controlling the end effector to pick up the target object after the target object is adjusted to the target pose; wherein the end effector is in motion during an entire phase from the end effector colliding with the target object to picking up the target object;

obtaining, based on a unified equation of state corresponding to a hybrid system comprising the end effector and the target object, pose information and force information of the hybrid system at each time step of the entire phase; and

controlling the end effector at each time step of the entire phase according to the pose information and the force information of the hybrid system at each time step of the entire phase.

11 . The apparatus according to claim 10 , wherein a position where the end effector collides with the target object is located at a bottom of the target object, to allow the target object to fall toward the end effector.

12 . The apparatus according to claim 11 , wherein the rotation control module is further configured to control the end effector to rotate in a direction opposite to a falling direction in a process of the target object falling over.

13 . The apparatus according to claim 10 , wherein a pick-up mode of the end effector is non-grasping type in which the target object is held by a palm part of the end effector, and a static friction force based on a bearing surface of the end effector allows the target object and the end effector to remain relatively stationary.

14 . The apparatus according to claim 13 , wherein the pick-up control module is further configured to control the end effector to rotate until the target object and the palm part of the end effector is relatively stationary.

15 . The apparatus according to claim 10 , wherein a pick-up mode of the end effector is grasping type in which a finger part of the end effector is closed toward a palm part of the end effector, to grasp the target object between the palm part and the finger part.

16 . The apparatus according to claim 10 , wherein the pose information and the force information of the hybrid system at each time step of the entire phase meet the following constraint conditions:

an inequality constraint condition, configured to constrain a friction force of a contact position between the end effector and the target object; and

an equality linkage constraint condition, configured to constrain a motion trajectory and an acting force between the end effector and the target object.

17 . The apparatus according to claim 16 , wherein the obtaining, based on the unified equation of state corresponding to the hybrid system comprising the end effector and the target object, the pose information and the force information of the hybrid system at each time step of the entire phase comprises:

constructing an objective function based on a Bellman optimal equation, the unified equation of state, and the constraint conditions; and

solving the objective function by using a primal dual augmented Lagrangian multiplier method, to obtain the pose information and the force information of the hybrid system at each time step of the entire phase.

18 . A non-transitory computer-readable storage medium, storing a computer program, the computer program being loaded and executed by a processor coupled to a robot to implement:

controlling an end effector of the robot to collide with a target object in a process of the end effector moving to the target object;

controlling the end effector to rotate relative to the target object, to adjust the target object to a target pose corresponding to a pick-up action;

controlling the end effector to pick up the target object after the target object is adjusted to the target pose; wherein the end effector is in motion during an entire phase from the end effector colliding with the target object to picking up the target object;

obtaining, based on a unified equation of state corresponding to a hybrid system comprising the end effector and the target object, pose information and force information of the hybrid system at each time step of the entire phase; and

controlling the end effector at each time step of the entire phase according to the pose information and the force information of the hybrid system at each time step of the entire phase.