Robot dynamic obstacle avoidance trajectory planning method, system and device and storage medium

Abstract

The invention discloses a robot dynamic obstacle avoidance trajectory planning method, which comprises the following steps: determining the position of a workpiece by taking a global camera as guidance, and acquiring a global path from a starting point to a target point by taking a Bi-RRT algorithm as a global planner in a motion planning process; when a moving obstacle is detected in the operation area, a depth camera on a robot manipulator is turned on to capture depth information of a local environment in real time; an obstacle in a local environment is simplified into an AABB bounding box, a mechanical arm connecting rod of the robot is simplified into space line segments, and whether the mechanical arm connecting rod collides with the obstacle or not is pre-judged by judging whether all the line segments intersect with the bounding box or not; and when the mechanical arm connecting rod collides with the obstacle, local motion planning is carried out based on the SAC algorithm, and a corresponding obstacle avoidance action instruction is generated. According to the method, the global optimal path can be quickly planned, and accurate dynamic obstacle avoidance can be realized. Correspondingly, the invention further provides a robot dynamic obstacle avoidance trajectory planning system, electronic equipment and a storage medium.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent motion planning for robotic arms, and in particular to a method, system, electronic equipment and storage medium for dynamic obstacle avoidance trajectory planning for a robot. Background Art

[0002] In modern manufacturing, the advancement of intelligent assembly and Industry 5.0 is driving a comprehensive upgrade of production methods. Intelligent assembly achieves efficient, precise and flexible production by integrating automation and information technology. Industry 5.0 emphasizes the deep collaboration between humans and intelligent robots, combining the creativity of human workers with the efficient production capacity of intelligent machines to achieve a more personalized and humanized production model. As the core equipment of intelligent assembly, the performance of robots directly affects the efficiency and reliability of the entire production system. In a complex and dynamic environment, robots face many obstacles and uncertainties, so real-time obstacle avoidance motion planning has become a key technology. Therefore, how to perceive environmental changes in real time, dynamically calculate the best obstacle avoidance path, and ensure the safety and continuity of the assembly process has become a difficult problem that needs to be solved urgently. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a robot dynamic obstacle avoidance trajectory planning method to achieve rapid planning of the global optimal path and accurate dynamic obstacle avoidance. Correspondingly, the present invention also provides a robot dynamic obstacle avoidance trajectory planning system, electronic device and storage medium.

[0004] According to one aspect of the present invention, a robot dynamic obstacle avoidance trajectory planning method is provided, comprising the following steps:

[0005] The workpiece position is determined with the global camera as a guide, and the Bi-RRT algorithm is used as the global planner in the motion planning process to obtain the global path from the starting point to the target point;

[0006] When a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time;

[0007] Simplify the obstacles in the local environment into AABB bounding boxes and simplify the robot's arm links into space segments and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box;

[0008] When the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves towards it;

[0009] If the robot deviates from the global path by more than a preset threshold during obstacle avoidance, a new global path is generated with the robot's current position as the new starting point.

[0010] In one embodiment of the present invention, the method of determining the position of the workpiece with the global camera as a guide and using the Bi-RRT algorithm as a global planner in the motion planning process to obtain a global path from the starting point to the target point includes:

[0011] First, starting point and target point As two root nodes, construct random trees respectively and ;

[0012] Then, perform collision detection calculation according to the step size ρ to find the distance sampling point The nearest node , if the node The path will not collide with obstacles, then the random tree Adding a new node ;

[0013] Then with the new node As the target point, the random tree According to the above algorithm, the new node Expand until two random trees and Connect and find the global path.

[0014] In one embodiment of the present invention, the obstacles in the local environment are simplified into AABB bounding boxes, and the robot arm links are simplified into space line segments. and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box, including:

[0015] By solving the space segment and The intersection with the AABB bounding box is calculated and ; When the judgment condition is When , the robot arm link will collide with the obstacle bounding box, otherwise no collision will occur;

[0016] in:

[0017]

[0018]

[0019] With robot arm connecting rod For example, the robot arm connecting rod Two planes in the obstacle bounding box and The intersection points are and , the calculation formula is as follows:

[0020]

[0021] in, and It is the robot arm connecting rod The two ends of the robot correspond to and When calculating the coefficients of the joint coordinates, we only need to find the x component:

[0022]

[0023] The result is:

[0024]

[0025] Define the robot arm links With two Y planes and The intersection points are and , Robotic arm connecting rod With two Z planes and The intersection points are and ;

[0026] Similarly, we can calculate:

[0027]

[0028]

[0029] in, and Respectively represent the minimum and maximum points of the AABB envelope.

[0030] In one embodiment of the present invention, when the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves toward it, including:

[0031] The SAC algorithm updates the policy and Q-value network by maximizing the objective function of expected reward and entropy;

[0032] During the training process, priority experience replay is used to measure the importance of samples through time difference error, and high-value samples are sampled first;

[0033] Design a reward function system: give negative rewards when the distance between the robot and the obstacle decreases, and give positive rewards when the distance increases; when in a safe area, the reward is proportional to the distance to the target; give a large negative reward when colliding, and a large positive reward when reaching the target.

[0034] In one embodiment of the present invention, if the robot deviates from the global path by more than a preset threshold during the obstacle avoidance process, a new global path is generated with the current position of the robot as a new starting point, including:

[0035] Determine whether the distance between the current robot end effector and the nearest node is greater than the set value , if it is greater than , the current position of the robot is taken as the new starting point, and the Bi-RRT algorithm is used to generate a new global path.

[0036] According to another aspect of the present invention, a robot dynamic obstacle avoidance trajectory planning system is provided, comprising:

[0037] Global path planning module: Use the global camera as a guide to determine the workpiece position, use the Bi-RRT algorithm as a global planner in the motion planning process, and obtain the global path from the starting point to the target point;

[0038] Detection module: When a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time;

[0039] Prediction module: Simplify obstacles in the local environment into AABB bounding boxes and simplify the robot's arm links into space segments and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box;

[0040] Local motion planning module: When the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves towards it;

[0041] Global path replanning module: If the robot deviates from the global path by more than a preset threshold during obstacle avoidance, a new global path is generated with the robot's current position as the new starting point.

[0042] According to another aspect of the present invention, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0043] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0044] processor;

[0045] A storage medium stores a computer program thereon, wherein the computer program executes the steps of the method described above when executed by the processor.

[0046] The present invention is a robot dynamic obstacle avoidance trajectory planning method that combines a traditional planning algorithm with a deep reinforcement learning algorithm. The bidirectional rapid exploration random tree (Bi-RRT) method is used as a global planner to quickly plan the global optimal path; then, considering the real-time nature of the assembly process, the soft actor critic algorithm (SAC) is used as a local obstacle avoider to accurately avoid obstacles, and in the planning process, the nearest node generated by the Bi-RRT is found and regarded as the target for local obstacle avoidance, thereby reducing the complexity of the model, greatly improving the assembly efficiency and automation of the robot, and significantly saving labor. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present invention, and together with the specification are used to explain the principles of the present invention. Obviously, the accompanying drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0048] Figure 1 It is a flow chart of a robot dynamic obstacle avoidance trajectory planning method in one embodiment of the present invention.

[0049] Figure 2 It is a module diagram of a robot dynamic obstacle avoidance trajectory planning system in one embodiment of the present invention.

[0050] Figure 3 It is a schematic diagram of the structure of an electronic device in one embodiment of the present invention.

[0051] Figure 4 It is a schematic diagram of the structure of a computer-readable storage medium in one embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the description of the present invention more detailed and complete, reference may be made to the accompanying drawings and various embodiments described below, in which the same numbers represent the same or similar components. On the other hand, well-known components and steps are not described in the embodiments to avoid unnecessary limitations on the present invention. In the embodiments and claims, the description of "connection" may generally refer to a component being indirectly coupled to another component through other components, or a component being directly connected to another component without passing through other components. In the embodiments and claims, the description of "online" may generally refer to a component being indirectly connected to another component through other components for wired and / or wireless communication, or a component being physically connected to another component without passing through other components. In the embodiments and claims, unless the article is specifically limited in the text, "a" and "the" may generally refer to a single or multiple. "About", "approximately" or "roughly" as used herein is used to modify any quantity that can be slightly changed, but such slight changes do not change its essence. In the embodiments, if there is no special explanation, the error range of the numerical value modified by "about", "approximately" or "substantially" is generally allowed within 20%, preferably within 10%, and more preferably within 5%.

[0053] According to one aspect of the present invention, a robot dynamic obstacle avoidance trajectory planning method is provided. Figure 1 is a flow chart of a method in one embodiment of the present invention. Figure 1 As shown, the method may include the following steps:

[0054] S1, using the global camera as a guide to determine the workpiece position, using the Bi-RRT algorithm as a global planner in the motion planning process to obtain the global path from the starting point to the target point. Specifically, the following steps may be included:

[0055] First, starting point and target point As two root nodes, construct random trees respectively and .

[0056] Then, perform collision detection calculation according to the step size ρ to find the distance sampling point The nearest node , if the node The path will not collide with obstacles, then the random tree Adding a new node .

[0057] Then with the new node As the target point, the random tree According to the above algorithm, the new node Expand until two random trees and Connect and find the global path.

[0058] S2, when a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time.

[0059] S3, simplify the obstacles in the local environment into AABB bounding boxes and simplify the robot's arm links into space segments and , by determining whether each line segment intersects with the bounding box, it is predicted whether the robot arm link will collide with an obstacle.

[0060] Specifically, by solving the space segment and The intersection with the AABB bounding box is calculated and ; When the judgment condition is , the robot arm link will collide with the obstacle bounding box, otherwise no collision will occur.

[0061] in:

[0062]

[0063]

[0064] With robot arm connecting rod For example, the robot arm connecting rod Two planes in the obstacle bounding box and The intersection points are and , the calculation formula is as follows:

[0065]

[0066] in, and It is the robot arm connecting rod The two ends of the robot correspond to and When calculating the coefficients of the joint coordinates, we only need to find the x component:

[0067]

[0068] The result is:

[0069]

[0070] Define the robot arm links With two Y planes and The intersection points are and , Robotic arm connecting rod With two Z planes and The intersection points are and .

[0071] Similarly, we can calculate:

[0072]

[0073]

[0074] in, and Respectively represent the minimum and maximum points of the AABB envelope.

[0075] S4, when the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves towards it.

[0076] Specifically, the SAC algorithm updates the strategy and Q-value network by maximizing the objective function of expected reward and entropy. The introduction of entropy can increase the randomness of the strategy, thereby improving the exploration ability and learning efficiency.

[0077] During the training process, in order to improve the efficiency of sample utilization, Prioritized Experience Replay (PER) can be used to measure the importance of samples through the time difference (TD) error, and give priority to sampling high-value samples; in order to reduce over-estimation and improve the stability of training, the SAC algorithm uses two Q networks and uses a smaller Q value for gradient calculation. In addition, SAC also contains a policy network , so it is necessary to consider the TD errors of the three networks simultaneously. TD error The metric chosen to evaluate the value of experience is calculated as follows:

[0078]

[0079] in, is a tuning parameter that balances the weight of the TD error calculated by the policy network and the Q network. To prevent the network from overfitting, experiences are extracted probabilistically to ensure that even experiences with 0 TD error are sampled. The priority value of each experience is defined as:

[0080]

[0081] in, , It represents a very small value, and the probability of preventing sampling is 0;

[0082] To ensure that the robot can effectively avoid obstacles during movement, a reward function system is designed: negative rewards are given when the distance between the robot and the obstacle decreases, and positive rewards are given when the distance increases; when in a safe area, the reward is proportional to the target distance; a large negative reward is given in the event of a collision, and a large positive reward is given when the target is reached.

[0083] Among them, the dynamic obstacle avoidance reward is calculated as follows:

[0084]

[0085] in, represents the corresponding weight, , It represents the distance between each link and the center point of the obstacle. Less than the previous When , it means that the distance between the robot and the dynamic obstacles in the environment is shrinking, and a negative reward is obtained; when Greater than When , it means that the distance between the robot and the dynamic obstacles in the environment is increasing, and the robot obtains a positive reward;

[0086] The safe area reward formula is calculated as follows:

[0087]

[0088] in, represents the corresponding weight, , is the distance between the target node and the end effector of the robot arm in the working area; when the robot arm is in the safe area, the reward is proportional to the target distance;

[0089] When any of the three robot links collides with a moving obstacle in the environment, a significant negative reward is generated. , when the robot reaches the target node, it will receive a considerable positive reward .

[0090] S5, if the robot deviates from the global path by more than a preset threshold during the obstacle avoidance process, a new global path is generated with the current position of the robot as a new starting point.

[0091] Specifically, determine whether the distance between the current robot end effector and the nearest node is greater than the set value If it is greater than, it means that the robot has deviated too far from the global path, and the global path planning is retriggered. The current position of the robot is used as the new starting point, and the Bi-RRT algorithm is used to recalculate and generate a new global path.

[0092] The present invention combines an industrial camera with a depth camera: placing an industrial camera at a position where the global perspective can be observed, and installing a depth camera on the robot's manipulator to achieve comprehensive perception of the global and local environments, which can improve perception accuracy and environmental understanding capabilities. The present invention implements local obstacle avoidance of the robot arm through the SAC (Soft Actor-Critic) algorithm, and global path planning of the robot arm through the Bi-RRT (Bidirectional Rapidly-exploring Random Tree) algorithm, and combines them to jointly plan motion planning in autonomous assembly tasks of the robot arm. The present invention uses obstacle point cloud data obtained by industrial cameras and depth cameras, simplifies it through AABB (axis-aligned bounding box), and then uses the AABB bounding box to perform collision detection on obstacles, which can improve computing efficiency and simplify obstacle avoidance tasks in complex environments.

[0093] According to another aspect of the present invention, a robot dynamic obstacle avoidance trajectory planning system is provided. Figure 2 FIG. 1 is a block diagram of a robot dynamic obstacle avoidance trajectory planning system in one embodiment of the present invention. Figure 2 As shown, the robot dynamic obstacle avoidance trajectory planning system 300 in this embodiment includes a global path planning module 310, a detection module 320, a prediction module 330, a local motion planning module 340, and a global path re-planning module 350. Based on the global path planning module 310, the position of the workpiece is determined with the guidance of the global camera, and the Bi-RRT algorithm is used as the global planner in the motion planning process to obtain the global path from the starting point to the target point. Based on the detection module 320, when a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time. Based on the prediction module 330, the obstacles in the local environment are simplified into AABB bounding boxes, and the robot's robotic arm links are simplified into spatial line segments. and , by judging whether each line segment intersects with the bounding box, it is predicted whether the robot arm connecting rod will collide with the obstacle. Based on the local motion planning module 340, when the robot arm connecting rod is about to collide with the obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves to it. Based on the global path replanning module 350, if the robot deviates from the global path by more than a preset threshold during the obstacle avoidance process, a new global path is generated with the current position of the robot as the new starting point. The present invention combines an industrial camera with a depth camera: an industrial camera is placed at a position where the global perspective can be observed, and a depth camera is installed on the robot's manipulator to achieve comprehensive perception of the global and local environment, which can improve the perception accuracy and environmental understanding ability. The present invention realizes local obstacle avoidance of the robot arm through the SAC (Soft Actor-Critic) algorithm, realizes global path planning of the robot arm through the Bi-RRT (Bidirectional Rapidly-exploring Random Tree) algorithm, and combines them to jointly plan the motion planning in the robot arm autonomous assembly task. The present invention utilizes the obstacle point cloud data acquired by industrial cameras and depth cameras, simplifies it through AABB (axis-aligned bounding box), and then uses the AABB bounding box to perform collision detection on the obstacles, which can improve the computing efficiency and simplify the obstacle avoidance task in complex environments.

[0094] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0095] processor;

[0096] A storage medium stores a computer program thereon, wherein the computer program executes the steps of the method described above when executed by the processor.

[0097] Refer to the following Figure 3 To describe an electronic device 600 according to an embodiment of the present application. Figure 3 The electronic device 600 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0098] like Figure 3 As shown, the electronic device 600 is in the form of a general computing device. The components of the device 600 may include, but are not limited to: at least one processor 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processor 610), a display unit 640, etc.

[0099] The storage unit stores program codes, which can be executed by the processor 610, so that the processor 610 executes the steps of various exemplary embodiments of the present application described in the electronic prescription circulation processing method section of this specification. For example, the processor 610 can execute the steps in the above method.

[0100] The storage unit 620 may include a storage medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0101] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0102] Bus 630 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

[0103] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), and may also communicate with one or more devices that enable tenants to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 650. In addition, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network 102 adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0104] The present invention combines an industrial camera with a depth camera: placing an industrial camera at a position where the global perspective can be observed, and installing a depth camera on the robot's manipulator to achieve comprehensive perception of the global and local environments, which can improve perception accuracy and environmental understanding capabilities. The present invention implements local obstacle avoidance of the robot arm through the SAC (Soft Actor-Critic) algorithm, and global path planning of the robot arm through the Bi-RRT (Bidirectional Rapidly-exploring Random Tree) algorithm, and combines them to jointly plan motion planning in autonomous assembly tasks of the robot arm. The present invention uses obstacle point cloud data obtained by industrial cameras and depth cameras, simplifies it through AABB (axis-aligned bounding box), and then uses the AABB bounding box to perform collision detection on obstacles, which can improve computing efficiency and simplify obstacle avoidance tasks in complex environments.

[0105] According to another aspect of the present invention, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are executed.

[0106] refer to Figure 4 In one embodiment, the program product 800 for implementing the above method may adopt a portable compact disk read-only memory (CD-ROM) and include program code, and may be run on a server. However, those skilled in the art should understand that the program product referred to in this application is not limited thereto, and the computer storage medium may be any tangible medium containing or storing a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

[0107] It should be noted that the present application can be implemented in software and / or a combination of software and hardware, for example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer or any other similar hardware device. In one embodiment, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer-readable recording medium, for example, a RAM memory, a magnetic or optical drive or a floppy disk and the like. In addition, some steps or functions of the present application can be implemented using hardware, for example, as a circuit that cooperates with a processor to perform various steps or functions.

[0108] In addition, a part of the present application may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can be called or provided through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer storage media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer storage medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0109] Communication media include media by which communication signals containing, for example, computer readable instructions, data structures, program modules or other data are transmitted from one system to another. Communication media may include guided transmission media (such as cables and wires, for example, fiber optic, coaxial, etc.) and wireless (unguided transmission) media that can propagate energy waves, such as acoustic, electromagnetic, RF, microwave and infrared. Computer readable instructions, data structures, program modules or other data may be embodied as a modulated data signal in, for example, a wireless medium (such as a carrier wave or similar mechanism such as embodied as part of spread spectrum technology). The term "modulated data signal" refers to a signal whose one or more characteristics are changed or set in such a manner as to encode information in the signal. Modulation may be analog, digital, or a hybrid modulation technique.

[0110] By way of example and not limitation, computer-readable storage media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules or other data. For example, computer-readable storage media include (but are not limited to) volatile memory, such as random access memory (RAM, DRAM, SRAM); and non-volatile memory, such as flash memory, various read-only memories (ROM, PROM, EPROM, EEPROM), magnetic and ferromagnetic / ferroelectric memories (MRAM, FeRAM); and magnetic and optical storage devices (hard disks, magnetic tapes, CDs, DVDs); or other media now known or later developed that can store computer-readable information or data for use by a computer system.

[0111] The present invention combines an industrial camera with a depth camera, places the industrial camera at a position where the global perspective can be observed, and installs a depth camera on the robot's manipulator to achieve comprehensive perception of the global and local environments, thereby improving perception accuracy and environmental understanding capabilities. The present invention implements local obstacle avoidance of the robot arm through the SAC (Soft Actor-Critic) algorithm, and global path planning of the robot arm through the Bi-RRT (Bidirectional Rapidly-exploring Random Tree) algorithm, and combines them to jointly plan motion planning in autonomous assembly tasks of the robot arm. The present invention uses obstacle point cloud data obtained by industrial cameras and depth cameras, simplifies it through AABB (axis-aligned bounding box), and then uses the AABB bounding box to perform collision detection on obstacles, which can improve computing efficiency and simplify obstacle avoidance tasks in complex environments.

[0112] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A robot dynamic obstacle avoidance trajectory planning method, characterized in that: include: The workpiece position is determined with the global camera as a guide, and the Bi-RRT algorithm is used as the global planner in the motion planning process to obtain the global path from the starting point to the target point; When a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time; Simplify the obstacles in the local environment into AABB bounding boxes and simplify the robot's arm links into space segments and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box; When the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves towards it; If the robot deviates from the global path by more than a preset threshold during obstacle avoidance, a new global path is generated with the robot's current position as the new starting point.

2. The method according to claim 1, characterized in that The method of determining the position of the workpiece with the global camera as a guide and using the Bi-RRT algorithm as a global planner in the motion planning process to obtain a global path from the starting point to the target point includes: First, starting from the and target point As two root nodes, construct random trees respectively and ; Then, perform collision detection calculation according to the step size ρ to find the distance sampling point The nearest node , if the node The path will not collide with obstacles, then the random tree Adding a new node ; Then with the new node As the target point, the random tree According to the above algorithm, the new node Expand until two random trees and Connect and find the global path.

3. The method according to claim 1, characterized in that The obstacles in the local environment are simplified into AABB bounding boxes, and the robot arm links are simplified into space line segments. and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box, including: By solving the space segment and The intersection with the AABB bounding box is calculated and ; When the judgment condition is When , the robot arm link will collide with the obstacle bounding box, otherwise no collision will occur; in: With robot arm connecting rod For example, the robot arm connecting rod Two planes in the obstacle bounding box and The intersection points are and , the calculation formula is as follows: in, and It is the robot arm connecting rod The two ends of the robot and When calculating the coefficients of the joint coordinates, find the x component: The result is: Define the robot arm links With two Y planes and The intersection points are and , Robotic arm connecting rod With two Z planes and The intersection points are , . Similarly, calculate: in, and Respectively represent the minimum and maximum points of the AABB envelope.

4. The method according to claim 1, characterized in that: When the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves toward it, including: The SAC algorithm updates the policy and Q-value network by maximizing the objective function of expected reward and entropy; During the training process, priority experience replay is used to measure the importance of samples through time difference error, and high-value samples are sampled first; Design a reward function system: give negative rewards when the distance between the robot and the obstacle decreases, and give positive rewards when the distance increases; when in a safe area, the reward is proportional to the distance to the target; give a large negative reward when colliding, and a large positive reward when reaching the target.

5. The method according to claim 1, characterized in that If the robot deviates from the global path by more than a preset threshold during the obstacle avoidance process, a new global path is generated with the current position of the robot as a new starting point, including: Determine whether the distance between the current robot end effector and the nearest node is greater than the set value , if it is greater than , the current position of the robot is taken as the new starting point, and the Bi-RRT algorithm is used to generate a new global path.

6. A robot dynamic obstacle avoidance trajectory planning system, characterized in that: include: Global path planning module: Use the global camera as a guide to determine the workpiece position, use the Bi-RRT algorithm as a global planner in the motion planning process, and obtain the global path from the starting point to the target point; Detection module: When a moving obstacle is detected in the working area, the depth camera on the robot manipulator is turned on to capture the depth information of the local environment in real time; Prediction module: Simplify obstacles in the local environment into AABB bounding boxes and simplify the robot's arm links into space segments and , predicting whether the robot arm link will collide with an obstacle by judging whether each line segment intersects with the bounding box; Local motion planning module: When the robot arm link is about to collide with an obstacle, local motion planning is performed based on the SAC algorithm, and corresponding obstacle avoidance action instructions are generated, so that the robot takes the nearest global path node as the local target point and moves towards it; Global path replanning module: If the robot deviates from the global path by more than a preset threshold during obstacle avoidance, a new global path is generated with the robot's current position as the new starting point.

7. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, performs the steps of the method according to any one of claims 1 to 5.

8. An electronic device, characterized in that: The electronic device comprises: processor; A storage medium having a computer program stored thereon, wherein the computer program, when executed by the processor, performs the steps of the method according to any one of claims 1 to 5.

Images