WO2023029901A1

WO2023029901A1 - Robot collision prediction method, computer storage medium, and electronic device

Info

Publication number: WO2023029901A1
Application number: PCT/CN2022/111071
Authority: WO
Inventors: 李明洋; 戚祯祥; 刘博峰; 邵威
Original assignee: 节卡机器人股份有限公司
Priority date: 2021-09-01
Filing date: 2022-08-09
Publication date: 2023-03-09
Also published as: CN113664832A

Abstract

A robot collision prediction method, a computer storage medium, and an electronic device. The robot collision prediction method comprises: monitoring a movement state of a robot; according to the movement state, predicting a future movement trajectory of the robot; and according to a size parameter of the robot, determining whether a collision with the robot will occur in the future movement trajectory.

Description

Robot collision prediction method, computer storage medium and electronic equipment

This application claims the priority of a Chinese patent application with application number 202111019083.5 filed with the China Patent Office on September 1, 2021, the entire contents of which are incorporated herein by reference.

technical field

The present application relates to the technical field of robots, for example, to a robot collision prediction method, computer storage media and electronic equipment.

Background technique

In the process of dragging and teaching the robot, controlling the robot through the software development kit (Software Development Kit, SDK), and adaptively moving the robot based on vision, the collision of the robot itself may occur.

For the safety of the robot body and the user's use of the robot, it is necessary to stop the robot and not continue to hit the body when the robot body collides with the body, which requires the robot to have the function of self-collision detection.

Related technologies rely on sensors or motors for collision detection of robots. For example, several commonly used collision detection methods:

(1) Use the wrist force sensor to detect collision: it can accurately detect the collision force at the end of the robot, but cannot detect the collision of other parts of the robot. The detection range is limited, and it is usually used for hand collisions such as grinding force and assembly force. force detection.

(2) Use electronic skin to detect collisions: the electronic skin covering the entire robot perception can detect collisions at any part of the robot. But the disadvantage is that the wiring is complicated and the anti-interference ability is poor.

(3) Collision detection by motor current or feedback torque: This is a collision scheme widely used in various industrial robots without using other sensors. Its advantage is that the detection range can cover the entire surface of the robot, but its disadvantages It is due to the friction problem at the joint that may make the detection inaccurate.

In the above methods, the collision signal can only be detected after the robot has collided. In this industry, there is no function related to the robot’s own collision warning and prompting. Since the robot can only stop or take evasive actions in time after receiving the collision signal, it cannot completely prevent the occurrence of the collision event, which will inevitably cause harm to personnel. Physical damage or damage to other objects.

Contents of the invention

This application provides a robot collision prediction method, computer storage media and electronic equipment to solve related technologies. Because the robot can only stop in time or make evasive actions after receiving the collision signal, it cannot completely prevent the occurrence of collision events, which will inevitably cause The problem of damage to people, damage to the robot itself, or damage to other objects.

A robot collision prediction method provided by the application includes:

Monitor the movement status of the robot;

According to the motion state, predict the future trajectory of the robot in real time;

According to the size parameters of the robot, it is judged whether the robot will collide in the future trajectory.

In some implementations, according to the size parameters of the robot, it is judged whether the robot collides in the future trajectory, including:

Simplify the mechanical model of the robot into multiple cylindrical bounding boxes, and select two cylindrical bounding boxes in the multiple cylindrical bounding boxes;

According to the size parameters of the two cylindrical bounding boxes, it is judged whether the two cylindrical bounding boxes collide in the future motion trajectory.

In the above technical solution, the mechanical model of the robot is simplified into a plurality of cylindrical bounding boxes, and two cylindrical bounding boxes are used as targets for judgment each time. When the robot has N (N>2) cylindrical bounding boxes, it needs to judge whether it collides with itself one or more times. For the calculation of the number of judgments, first select any two from the N cylindrical bounding boxes each time. Different cylindrical bounding boxes in no order, a total of N×(N-1)/2 times, and then subtract the adjacent cylindrical bounding boxes that will not collide (ie N-1 times), and finally can be manually Screening simplifies the number of judgments.

In some implementations, judging whether two cylindrical bounding boxes collide in the future trajectory includes:

If the distance between the two geometric center points of the two cylindrical bounding boxes is lower than a preset threshold, it is determined that the robot collides.

In some implementation manners, judging whether two cylindrical bounding boxes collide in the future trajectory also includes:

If the distance between the two center points is greater than or equal to a preset threshold and the two end circles of the two cylindrical bounding boxes intersect, it is determined that the robot collides.

If the distance between the two center points is greater than or equal to the preset threshold, the two end circles of the two cylindrical bounding boxes do not intersect, the two central axes of the two cylindrical bounding boxes are coplanar and perpendicular, and the two central axes intersect , then it is determined that the robot has collided.

If the distance between the two center points is greater than or equal to the preset threshold, the two end circles of the two cylindrical bounding boxes do not intersect, the two central axes of the two cylindrical bounding boxes are parallel, and the axis distance between the two central axes is less than two The sum of the radii of the cylinders, and the projection of the two central axes on the same straight line overlaps, then it is determined that the robot has collided.

If the distance between the two center points is greater than or equal to the preset threshold, the end circles of the two cylindrical bounding boxes do not intersect, the two central axes of the two cylindrical bounding boxes are non-parallel and non-coplanar, and the two cylindrical bounding boxes If the corresponding two bus lines intersect or the length of the common perpendicular between the bus line of any cylindrical bounding box and the central axis of another cylindrical bounding box is less than the radius of the other cylindrical bounding box, it is determined that the robot has collided. Wherein, the generatrix is a generatrix on a cylindrical bounding box closest to another cylindrical bounding box.

In some embodiments, when the robot is in the first motion mode, controlling the joint rotation of the robot;

The motion state includes the current joint angle and the angular velocity of the joint;

According to the motion state, predict the future trajectory of the robot, including:

According to the current joint angle and the angular velocity of the joint, predict the future trajectory.

In the above technical solution, the first movement mode is to realize the movement of the robot by controlling the rotation of the joints of the robot, including the rotation speed of the joints and the clockwise or counterclockwise direction. In the first motion mode, according to the current joint angular position + the speed of a single joint × (time interval × number), the joint angular position of the robot after multiple time intervals can be obtained, which corresponds to the spatial position of the cylinder at multiple future moments , when the time interval is small enough, the future trajectory of the robot can be obtained.

In some embodiments, directly controlling the movement of the end point of the robot while the robot is in the second motion mode;

The motion state includes the pose of the current end point and the motion speed of the end point;

Predict the trajectory of the end point according to the pose of the current end point and the movement speed of the end point;

According to the movement trajectory of the end point, the change of the joint angle of the robot is inversely solved, and the future movement trajectory of the robot is obtained according to the change of the joint angle of the robot.

In the above technical solution, the second movement mode is to realize the movement of the robot by controlling the movement of the end point of the robot in Cartesian space. In the second motion mode, according to the current terminal pose + the movement speed of the terminal point × (time interval × number), the terminal poses of multiple time intervals can be obtained, and the inverse solutions can be obtained at multiple times The spatial position of the cylinder, when the time interval is small enough, the future trajectory of the robot can be obtained.

In some embodiments, when the robot is in the third motion mode, the robot is controlled to move according to the completed job programming, and the motion state of the robot is displayed synchronously in the image;

According to the motion state of the robot presented in the image, the future trajectory is predicted through an interpolation algorithm.

In the above technical solution, the third motion mode is to use the job programming of the robot to make the robot move according to the job programming, and to synchronize the motion process of the robot to the image for display. For the robot in the image, through the interpolation algorithm, Get the future trajectory of the robot.

A computer-readable storage medium provided by the present application, computer program instructions are stored on the computer-readable storage medium, and when the computer program instructions are read and executed by a processor, the robot collision prediction method as described above is executed .

An electronic device provided by the present application includes a memory and a processor, and computer program instructions are stored in the memory. When the computer program instructions are read and run by the processor, the robot collision prediction described in the above items is performed. method.

Description of drawings

FIG. 1 is a flow chart of a robot collision prediction method provided in an embodiment of the present application;

Fig. 2 is a perspective view of a robot provided by the embodiment of the present application;

Fig. 3 is a top view of a robot provided in the embodiment of the present application;

Fig. 4 is a flow chart of a robot collision judgment provided by the embodiment of the present application;

FIG. 5 is a structural diagram of an electronic device provided in an embodiment of the present application;

Fig. 6 is a schematic diagram of a busbar provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

The collision detection of the robot all depends on the sensor or the motor, and the robot can stop in time or take an evasive action after receiving the collision signal. A robot collision prediction method, computer storage medium, and electronic equipment provided in the embodiments of the present application provide an early warning mechanism for robot collisions, and extract feedback collision signals without relying on sensors and motors to avoid collisions.

Please refer to Fig. 1, Fig. 1 is a flow chart of a robot collision prediction method provided by the embodiment of the present application, including:

101. Monitor the movement state of the robot.

The robot collision prediction method of the embodiment of the present application is applicable to the movement modes of the robots including but not limited to the following three types:

In the first motion mode, the controller realizes the overall motion of the robot by controlling the rotation of multiple joints of the robot, including the rotation speed and clockwise or counterclockwise rotation of the multiple joints.

In the second movement mode, the controller realizes the movement of the robot by controlling the movement of the end point of the robot in Cartesian space. Among them, after the motion trajectory of the robot end point is determined, the unique motion process of the robot is determined through the optimal solution, that is, the motion trajectory of the robot end point corresponds to the motion process of the robot one by one, and the robot can be inversely solved by the motion trajectory of the robot end point. movement process.

In the third motion mode, the controller makes the robot move according to the job programming through the completed job programming, and synchronously displays the motion process of the robot on the image.

102. According to the motion state, predict the future motion trajectory of the robot in real time.

Corresponding to the three motion modes of the robot, they are described as follows:

In the first motion mode, the robot controls the joint rotation of the robot; the motion state includes the current joint angle and the angular velocity of the joint; according to the motion state, the future trajectory of the robot is predicted, including: predicting the future motion according to the current joint angle and the angular velocity of the joint track. In the first motion mode, according to the current joint angular position + the speed of a single joint × (time interval × number), the joint angular position of the robot after multiple time intervals can be obtained, which corresponds to the spatial position of the cylinder at multiple future moments , when the time interval is small enough, the future trajectory of the robot can be obtained.

In the second motion mode, the robot directly controls the movement of the end point of the robot; the motion state includes the pose of the current end point and the movement speed of the end point; according to the motion state, the future trajectory of the robot is predicted, including: according to the current end point The pose and the movement speed of the end point predict the movement trajectory of the end point; according to the movement trajectory of the end point, the change of the robot joint angle is inversely solved to obtain the future movement trajectory of the robot. The second movement mode is to realize the movement of the robot by controlling the movement of the end point of the robot in Cartesian space, that is, according to the pose of the current end point + the movement speed of the end point × (time interval × number), it can be The terminal poses of multiple time intervals are obtained, and the spatial position of the cylinder at multiple times is obtained after inverse solution. When the time interval is small enough, the future trajectory of the robot can be obtained.

In the third movement mode of the robot, the robot moves according to the completed job programming, and the movement state of the robot is displayed synchronously in the image. At this time, according to the movement state of the robot presented in the image, the future is predicted through an interpolation algorithm motion track. The third movement mode is to use the job programming of the robot to make the robot move according to the job programming, and to synchronize the movement process of the robot to the image for display, and to obtain the future movement of the robot through the interpolation algorithm for the robot in the image track.

103. According to the size parameters of the robot, it is judged whether the robot collides in the future trajectory.

To judge whether the robot collides with itself at any place, it is necessary to judge whether the multiple components of the robot collide. Therefore, the mechanical model of the robot is simplified into a plurality of cylindrical bounding boxes, each time surrounded by two cylinders The box is judged as the target. After multiple judgments, if there is no collision between any two cylindrical bounding boxes of the robot, it is considered that the robot does not collide with itself in the future trajectory. Although the embodiment of the present application discusses whether the robot collides with itself, the robot collision prediction method in the embodiment of the present application is still applicable to the prediction that the robot may collide with the external static environment or dynamic objects. For the external static environment, only a similar method is required. Import the parameters of objects in the environment in advance, simplify the objects in the environment into cuboid bounding boxes, and then add them to the judgment of robot collision prediction. For dynamic objects, refer to the cylindrical bounding box of the robot, simplify it into a cylindrical bounding box, and use it as a new cylindrical bounding box of the robot to add it to the judgment of robot collision prediction.

When the robot has N (N>2) cylindrical bounding boxes, it needs to judge whether it collides with itself one or more times. For the calculation of the number of judgments, first select any two from the N cylindrical bounding boxes each time. Different cylindrical bounding boxes in no order, a total of N×(N-1)/2 times, and then subtract the adjacent cylindrical bounding boxes that will not collide (ie N-1 times), and finally can be manually Screening simplifies the number of judgments. For example, please refer to the robot in Figure 2 and Figure 3, the robot has 9 cylindrical bounding boxes (a1, a2, a3, a4, a5, a6, b1, b2, b3), if two cylindrical bounding boxes The detection requires 9×8/2=36 detections, but due to the limitation of the mechanical structure, it is impossible for some structures to collide. For example, if two adjacent cylinder bounding boxes do not collide, then the 36-8=28 times, and then perform artificial judgment screening, and finally set 14 detections, as shown in the table below, Y indicates that the corresponding numbered cylindrical bounding box should be tested, and N indicates that no detection is required.

Table 1 The cylinder bounding box can undergo collision screening

For any two collidable cylindrical bounding boxes screened out above, according to the size parameters (including radius and height) of the two cylindrical bounding boxes, it is judged whether the two cylindrical bounding boxes collide in the future trajectory, Carry out the following judgment process:

Please refer to Figure 4. Figure 4 is a flow chart for judging whether two cylindrical bounding boxes collide in the future motion trajectory in the embodiment of the present application. The collision detection follows the following three processes, center distance pre-detection, end face intersection detection , side intersection detection sequence.

S1. Pre-detection of the center distance: if the line connecting the two cylindrical bounding boxes is short enough, it is determined that the robot has collided; if the line connecting the two cylindrical bounding boxes is not short enough, proceed to the next step of detection. Among them, the short enough connecting center line of two cylindrical bounding boxes may mean that the connecting center line of two cylindrical bounding boxes is lower than the preset threshold, and the not short enough connecting center line of two cylindrical bounding boxes may mean that two The centerline of the cylindrical bounding box is higher than or equal to the preset threshold.

S2. End face intersection detection: If the two end face circles of the two cylindrical bounding boxes intersect, it is determined that the robot has collided; if the two end face circles of the two cylindrical bounding boxes do not intersect, proceed to the next step.

S3. Side collision detection: Find the positional relationship between the two central axes of the two cylindrical bounding boxes.

(1) When the two central axes of the two cylindrical bounding boxes are coplanar and perpendicular, if the two central axes of the two cylindrical bounding boxes intersect with a T or a cross, it is determined that the robot collides; if the two cylindrical bounding boxes If the two central axes are not T-shaped and do not intersect with each other, it is determined that the robot does not collide.

(2) When the two central axes of the two cylindrical bounding boxes are parallel, if the axis distance is less than the sum of the radii of the two cylinders, and the projections of the two central axes of the two cylindrical bounding boxes on the same straight line overlap, then determine When the robot collides, if the wheelbase is not less than the sum of the radii of the two cylinders, or the projections of the two central axes of the bounding boxes of the two cylinders on the same straight line do not overlap, then it is determined that the robot does not collide.

(3) When the two central axes of the two cylindrical bounding boxes are non-parallel and non-coplanar, respectively find the generatrices closest to each other on each cylindrical bounding box, if the two generatrices intersect or the two cylindrical bounding boxes are close enough then It is determined that the robot collides, and if the bus lines do not intersect and the two cylindrical bounding boxes are not close enough, then it is determined that the robot does not collide. As shown in Figure 6, to find a generatrix on cylinder A closest to cylinder B, first assume that the center of cylinder B is P, project P onto the plane where the upper surface of cylinder A is located to obtain point P', and connect the center C _A and P' to get a radius vector r', and translate the axis X _A of cylinder A along the r' direction for a radius distance to get the generatrix, which is the closest generatrix between cylinder A and cylinder B. Assuming the generatrix is G _A , the length d _AB of the common perpendicular from G _A to the central axis X _B of cylinder B can be obtained. Similarly, a generatrix closest to cylinder A can also be found on cylinder B. Assuming the generatrix is G _B , the length d _BA of the common perpendicular from G _B to the central axis X _A of cylinder A can be obtained. If d _AB is less than or equal to the radius r _B of cylinder B, and/or d _BA is less than or equal to the radius r A of cylinder _A , it is considered that the distance between cylinder A and cylinder B is too close, and it is determined that the robot has collided. Alternatively, if the bus G _A intersects the bus G _B , it is determined that the robot has collided.

The robot collision prediction method of the embodiment of the present application can no longer rely on sensors when detecting whether the robot body collides with the body or the robot body collides with a known static environment. When detecting whether the robot body collides with the external dynamic environment or personnel, it is necessary to introduce sensors to obtain the motion status of the external dynamic environment or personnel.

Since every mechanical parameter of the robot is known, a relatively accurate cylindrical bounding box model of each joint part of the robot can be established. For example, use OpenGL to draw the robot model: when the robot body is moving, this function module will detect in real time whether some cylindrical bounding boxes on the predicted trajectory collide with each other. And this only depends on the software algorithm and the joint data acquired on the robot, without adding additional sensors. In addition, when it is necessary to detect whether the robot collides with the external static environment, a similar method can be adopted to import the parameters of other objects in the environment into this function module in advance, simplify the environmental objects into cuboid bounding boxes, and then add them to the robot collision detection.

Its effect is similar to the collision detection between objects in 3D (3Dimension, 3D) games, and the complexity of the detection algorithm can be appropriately adjusted according to the hardware capability to obtain faster detection and response speed.

The embodiment of the present application can also set the level of collision warning according to user needs, that is, if a collision occurs after 0.5s or 1s, the collision warning and alarm signal will be returned. Remind the operator in advance whether the ongoing operation is in compliance with the regulations and there is danger, so as to avoid the collision of the robot.

This method relies on the kinematics of the robot. After obtaining the velocity through the angle of the historical joints, it can predict the position of each cylindrical bounding box in the next period of time, and then detect whether each prediction time is collided according to the bounding box collision detection algorithm, and return the collision sign.

The collision detection scheme can not only make the detection range cover the entire surface of the robot body, the robot and the external environment, but also provide an early warning before the collision event occurs, which solves the problem of relying heavily on sensors in the collision detection technology scheme, the detection accuracy is not high enough, and the collision after collision In order to get feedback and other issues.

To sum up, the robot collision prediction method includes monitoring the motion state of the robot and predicting its future trajectory, combined with the size parameters of the robot to predict whether the robot will collide, and provides an early warning mechanism for the robot's own collision, without relying on sensors. The purpose is to The collision signal is fed back in advance to remind the user of dangerous operations, thereby avoiding collisions, improving the safety of the robot, and lowering the user's threshold for use.

FIG. 5 is a structural diagram of an electronic device provided by an embodiment of the present application. 5, the electronic device 500 includes: a processor 510, a memory 520, a communication interface 530, and an image collector 540. These components are interconnected and communicate with each other through a communication bus 550 and/or other forms of connection mechanisms (not shown).

Memory 520 includes one or more (only one is shown in the figure), which can be, but not limited to, random access memory (Random Access Memory, RAM), read only memory (Read Only Memory, ROM), programmable only Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM) )wait. Processor 510 and possibly other components may access memory 520 to read and/or write data therein.

The processor 510 includes one or more (only one is shown in the figure), which may be an integrated circuit chip, and has a signal processing capability. The above-mentioned processor 510 can be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a micro control unit (Micro Controller Unit, MCU), a network processor (Network Processor, NP) or other conventional processors; It can be a special-purpose processor, including a neural network processor (Neural-network Processing Unit, NPU), a graphics processor (Graphics Processing Unit, GPU), a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuits, ASIC), Field Programmable Gate Array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Also, when there are multiple processors 510, some of them may be general-purpose processors, and the other part may be special-purpose processors.

The communication interface 530 includes one or more (only one is shown in the figure), which can be configured to directly or indirectly communicate with other devices for data exchange. The communication interface 530 may include an interface for wired and/or wireless communication.

The image collector 540 includes one or more (only one is shown in the figure), which can be configured to collect images and send the collected images to the memory 520 for storage, and the processor 510 can process the images. The image collector 540 may be a camera (including components such as a lens and an image sensor).

One or more computer program instructions may be stored in the memory 520, and the processor 510 may read and execute these computer program instructions to implement the robot collision prediction method provided in the embodiment of the present application.

The structure shown in FIG. 5 is only for illustration, and the electronic device 500 may also include more or fewer components than those shown in FIG. 5 , or have a different structure from that shown in FIG. 5 . Various components shown in FIG. 5 may be implemented in hardware, software, or a combination thereof. The electronic device 500 may be a physical device, such as a personal computer (Personal Computer, PC), a notebook computer, a tablet computer, a mobile phone, a server, an embedded device, etc., or a virtual device, such as a virtual machine, a virtualized container, and the like. Moreover, the electronic device 500 is not limited to a single device, and may also be a combination of multiple devices or a cluster formed by a large number of devices.

For example, the electronic device 500 mentioned in the embodiment of the present application to respectively realize robot collision prediction with the first motion mode, the second motion mode and the third motion mode can be realized by the structure in FIG. 5 , but the first motion mode The electronic device 500 for implementing robot collision prediction, the second motion mode and the third motion mode may not necessarily include the image collector 540 in FIG. 5 during implementation.

The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores computer program instructions, and when the computer program instructions are read and run by the processor of the computer, the computer program instructions provided by the embodiments of the present application are executed. Robot collision prediction method. For example, the computer-readable storage medium can be implemented as the memory 520 in the electronic device 500 in FIG. 5 .

In the embodiments provided in this application, the disclosed devices and methods may be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

In addition, a unit described as a separate component may or may not be physically separated, and a component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

Furthermore, multiple functional modules in the embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

In this document, relational terms such as first and second etc. are used only to distinguish one entity or operation from another without necessarily requiring or implying any such relationship between these entities or operations. Actual relationship or sequence.

Claims

A robot collision prediction method, comprising:

monitoring the motion state of the robot;

Predicting the future trajectory of the robot in real time according to the motion state;

According to the size parameters of the robot, it is judged whether the robot collides in the future motion trajectory.
The method according to claim 1, wherein the judging whether the robot collides in the future trajectory according to the size parameters of the robot comprises:

Simplifying the mechanical model of the robot into a plurality of cylindrical bounding boxes, selecting two cylindrical bounding boxes in the plurality of cylindrical bounding boxes;

According to the size parameters of the two cylindrical bounding boxes, it is judged whether the two cylindrical bounding boxes collide in the future motion trajectory.
The method according to claim 2, wherein said judging whether said two cylindrical bounding boxes collide in said future trajectory comprises:

In response to a distance between two center points of the two cylindrical bounding boxes being lower than a preset threshold, it is determined that the robot collides.
The method according to claim 3, wherein said judging whether said two cylindrical bounding boxes collide in said future trajectory further comprises:

In response to the distance between the two center points being greater than or equal to the preset threshold and the two end circles of the two cylindrical bounding boxes intersecting, it is determined that the robot collides.
The method according to claim 3, wherein said judging whether said two cylindrical bounding boxes collide in said future trajectory further comprises:

In response to the distance between the two center points being greater than or equal to the preset threshold, the two end circles of the two cylindrical bounding boxes do not intersect, and the two central axes of the two cylindrical bounding boxes are coplanar and vertical, and the two central axes intersect, it is determined that the robot has collided.
The method according to claim 3, wherein said judging whether said two cylindrical bounding boxes collide in said future trajectory further comprises:

In response to the distance between the two center points being greater than or equal to the preset threshold, the two end circles of the two cylindrical bounding boxes do not intersect, and the two central axes of the two cylindrical bounding boxes are parallel, The wheelbase of the two central axes is less than the sum of the radii of the two cylinders, and the projections of the two central axes on the same straight line overlap, so it is determined that the robot has collided.
The method according to claim 3, wherein said judging whether said two cylindrical bounding boxes collide in said future trajectory further comprises:

In response to the distance between the two center points being greater than or equal to the preset threshold, the end circles of the two cylindrical bounding boxes do not intersect, and the two central axes of the two cylindrical bounding boxes are non-parallel and non-parallel Coplanar, the two generatrices corresponding to the two cylindrical bounding boxes intersect or the length of the common perpendicular between the generatrix of one cylindrical bounding box and the central axis of the other cylindrical bounding box is less than or equal to the other cylinder The radius of the volume bounding box determines that the robot collides, wherein the generatrix is a generatrix on a cylinder bounding box that is closest to another cylinder bounding box.
The method according to any one of claims 1-7, further comprising: controlling the joint rotation of the robot when the robot is in the first motion mode;

The motion state includes joint angles and joint angular velocities;

The predicting the future trajectory of the robot according to the motion state includes:

Predict the future motion trajectory according to the joint angle and the angular velocity of the joint.
The method according to any one of claims 1-7, further comprising, when the robot is in the second movement mode, directly controlling the movement of the end point of the robot;

The motion state includes the posture of the end point and the motion speed of the end point;

The predicting the future trajectory of the robot according to the motion state includes:

Predicting the motion track of the end point according to the pose of the end point and the motion speed of the end point;

According to the motion trajectory of the end point, the joint angle change of the robot is inversely solved, and the future motion trajectory of the robot is obtained according to the joint angle change of the robot.
The method according to any one of claims 1-7, further comprising: when the robot is in the third motion mode, controlling the robot to move according to the completed job programming, and synchronously displaying the robot in the image state of motion;

The predicting the future trajectory of the robot according to the motion state includes:

According to the motion state of the robot presented in the image, the future motion trajectory is predicted through an interpolation algorithm.
A computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the robot collision prediction method according to any one of claims 1-10 is executed.
An electronic device, comprising a memory and a processor, wherein computer program instructions are stored in the memory, and when the computer program instructions are read and executed by the processor, the method described in any one of claims 1-10 is executed. Robot collision prediction method.