Method for avoiding movement of a robot hand
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENYANG XINSONG SEMICON EQUIP CO LTD
- Filing Date
- 2026-07-03
- Publication Date
- 2026-08-04
AI Technical Summary
然而,这些方法普遍存在运算量大、实时性要求高的问题,且多为针对固定场景定制化设计——要么在程序内固化运动空间,要么预设离线运动轨迹
[0022] This invention addresses scenarios where robotic arms experience motion interference within their workspace, providing a universally applicable and highly efficient obstacle avoidance trajectory planning scheme. Compared to existing technologies that require pre-defined motion spaces or customized obstacle avoidance programs, this invention does not restrict the user's workspace, does not require the user to fix the control scheme, and always executes the optimal obstacle avoidance trajectory, effectively solving the problems of complex operation and poor scalability of traditional solutions.
Smart Images

Figure CN122500733A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of robotic arm technology, and particularly relates to a method for avoidance movement of a robotic arm. Background Technology
[0002] As a core execution device in the field of industrial automation, robotic arms are widely used in semiconductor manufacturing, precision assembly, material handling, and other applications. In practical applications, robotic arms are usually installed in the user's operating room. However, under certain working conditions, there may be motion interference between the user's operating platform and the robotic arm itself, preventing the robotic arm from directly rotating to the designated position. In such cases, the robotic arm needs to bypass the interfering operating platform to safely move into place.
[0003] Existing technologies have proposed several obstacle avoidance solutions, such as real-time obstacle detection and trajectory correction, or the generation of collision-free paths using random tree connection algorithms. Some solutions also assess risk by constructing a unified 3D coordinate system and calculating interference probabilities. However, these methods generally suffer from high computational complexity and real-time requirements, and are mostly customized designs for fixed scenarios—either fixing the motion space within the program or pre-setting offline motion trajectories. This approach is not only complex and inefficient but also lacks scalability. When the user's workstation layout changes or the robot needs to adapt to different chambers, it often requires re-teaching or modification of the control program, making it difficult to meet the flexible and ever-changing needs of industrial production. Furthermore, existing technologies cannot provide an efficient solution for the interference avoidance problem between cylindrical coordinate robot arms and multiple workstation disks within a limited chamber space.
[0004] Therefore, there is an urgent need for a universal, efficient, and automatically optimal obstacle avoidance solution. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a method for obstacle avoidance movement of a robotic arm.
[0006] A method for obstacle avoidance movement of a robotic arm includes the following steps:
[0007] At least one avoidance zone is established in the workspace of the robot arm, and the avoidance zone is jointly defined by the angular range in the direction of the rotation axis and the height range in the direction of the lifting axis.
[0008] During the process of the robotic arm moving directly from its current position to the target position, it is determined whether there is an intersection between the angle range traversed by the rotation axis and the angle range of the avoidance zone, and whether there is an intersection between the height range traversed by the lifting axis and the height range of the avoidance zone.
[0009] If both dimensions intersect, interference is determined, and an avoidance trajectory passing through at least one safety line is planned so that the robot arm can move around the avoidance zone to the target position.
[0010] The safety line includes a longitudinal safety line and / or a transverse safety line; the longitudinal safety line is a trajectory line with a fixed rotation axis angle and a variable lifting axis height; the transverse safety line is a trajectory line with a fixed lifting axis height and a variable rotation axis angle.
[0011] The height range of the lifting axis of the longitudinal safety line and the angle range of the rotation axis of the transverse safety line are set by the user according to the actual workspace.
[0012] If, during the process of the robotic arm moving directly from its current position to the target position, only the angle range traversed by the rotation axis intersects with the angle range of the avoidance zone, while the height range traversed by the lifting axis does not intersect with the height range of the avoidance zone, or only the height range traversed by the lifting axis intersects with the height range of the avoidance zone, while the angle range traversed by the rotation axis does not intersect with the angle range of the avoidance zone, then it is determined as non-interference.
[0013] The avoidance zone is a cuboid space region defined by two diagonal points, which correspond to two boundary values in the direction of rotation axis and two boundary values in the direction of elevation axis, respectively.
[0014] It also includes: when there are multiple feasible avoidance trajectories, calculating the total distance of each trajectory and selecting the trajectory with the shortest total distance to execute.
[0015] When multiple feasible avoidance trajectories involve parallel safety lines, the comparison of total travel distances simplifies to:
[0016] Calculate the first sum and the second sum. The first sum is the sum of the rotation distance from the current point to the first safety point and the rotation distance from the second safety point to the target point. The second sum is the sum of the rotation distance from the current point to the third safety point and the rotation distance from the fourth safety point to the target point.
[0017] Compare the first sum with the second sum, and select the trajectory with the smaller sum to execute.
[0018] The avoidance trajectory includes: moving from the current position to a first position on the safety line, moving along the safety line to a second position, and then moving from the second position to the target position.
[0019] The workspace contains multiple avoidance zones, each corresponding to a different spatial interference region.
[0020] The decision-making process is performed automatically before each motion task is executed by the robotic arm.
[0021] By employing the above technical solution, the present invention has at least the following beneficial effects:
[0022] This invention addresses scenarios where robotic arms experience motion interference within their workspace, providing a universally applicable and highly efficient obstacle avoidance trajectory planning scheme. Compared to existing technologies that require pre-defined motion spaces or customized obstacle avoidance programs, this invention does not restrict the user's workspace, does not require the user to fix the control scheme, and always executes the optimal obstacle avoidance trajectory, effectively solving the problems of complex operation and poor scalability of traditional solutions.
[0023] This invention models the interfering rigid body in the workspace and designs a safety line to assist in completing the avoidance trajectory. It can flexibly adapt to various application scenarios with interference conditions, without the need to redevelop control programs for different chamber environments, and has good portability and scalability.
[0024] This invention uses a path selection algorithm to automatically select the shortest running path when there are multiple avoidance trajectories, thus keeping the robot arm's handling efficiency at its highest level. This solves the problem that existing solutions cannot always maintain the optimal path planning motion in complex handling operations. Attached Figure Description
[0025] Figure 1 This is a schematic diagram of the layout of the robotic arm and the interference chamber in an embodiment of the present invention (wherein, the rectangular box in the figure indicates the elbow joint of the robotic arm).
[0026] Figure 2 This is a schematic diagram showing the distribution of each workstation in the chamber in an embodiment of the present invention;
[0027] Figure 3 This is a schematic diagram illustrating the principle of the avoidance motion trajectory in an embodiment of the present invention;
[0028] Figure 4 This is a schematic diagram of the avoidance zone modeling in an embodiment of the present invention (where dT represents the angular span of the avoidance zone in the direction of the rotation axis, and dZ represents the height span of the avoidance zone in the direction of the lifting axis).
[0029] Figure 5 This is a schematic diagram illustrating a scenario where there is interference in the avoidance zone effectiveness determination logic in an embodiment of the present invention;
[0030] Figure 6 This is a schematic diagram illustrating a scenario in an embodiment of the present invention where only the rotation angle range of the robotic arm and the angle range of the avoidance range intersect;
[0031] Figure 7 This is a schematic diagram illustrating a scenario in an embodiment of the present invention where only the lifting height range of the robotic arm and the height range of the avoidance range overlap.
[0032] Figure 8 This is a schematic diagram of the longitudinal and transverse safety lines in an embodiment of the present invention;
[0033] Figure 9 This is a schematic diagram of multi-path selection comparison in an embodiment of the present invention;
[0034] Figure 10 This is a schematic diagram illustrating the application of an embodiment of the present invention in a specific semiconductor chamber. Detailed Implementation
[0035] To better explain and facilitate understanding of the present invention, the technical solution and effects of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
[0036] This embodiment uses a cluster-type device in semiconductor manufacturing equipment as an example for illustration. Figures 1-2 As shown, LLA and LLB are loading interlocking chambers, i.e., user operating platforms, shaped like disks, used to transition the wafer between atmospheric and vacuum environments; PM1 and PM2 are process modules used to perform etching, deposition, and other processing on the wafer; Buffer is a buffer station used for temporary wafer storage; and ROB is the robot arm itself. LLA is located above LLB, while PM1, PM2, and the Buffer station are located below LLB. In this spatial layout, when the robot arm moves from the bottom layer to a station directly above LLA or LLB—for example, when the robot arm moves from the Buffer station to a station directly above LLA—if the robot arm rotates directly horizontally, the robot arm itself will interfere with and collide with LLA. Furthermore, when the robot arm rotates clockwise to a specific angle, the robot arm's elbow joint will also interfere with LLA. Therefore, it is necessary to plan the robot arm's movement trajectory to avoid collisions.
[0037] Combination Figures 3-4As shown, the avoidance zone is first modeled. In this embodiment, the robot uses a cylindrical coordinate system for motion, which includes rotational motion around the rotation axis (T-axis) and vertical motion along the lifting axis (Z-axis). Based on this, the avoidance method provided in this embodiment constructs each avoidance zone (Block) as a cuboid spatial region. The spatial range of the entire avoidance zone can be determined by the two diagonal points of the cuboid spatial region. Specifically, each avoidance zone (Block) is defined by two boundary values in the T-axis direction, denoted as the first rotational boundary value BlockPoint1.T and the second rotational boundary value BlockPoint2.T, respectively, and satisfying BlockPoint1.T ≤ BlockPoint2.T; similarly, it is defined by two boundary values in the Z-axis direction, denoted as the first vertical lifting boundary value BlockPoint1.Z and the second vertical lifting boundary value BlockPoint2.Z, respectively, and satisfying BlockPoint1.Z ≤ BlockPoint2.Z. Each avoidance zone is defined by parameters in two dimensions: the angle range is defined by the two boundary values BlockPoint1.T and BlockPoint2.T of the angle range through the rotation axis, and the height range is defined by the two boundary values BlockPoint1.Z and BlockPoint2.Z of the height range through the lifting axis, thus forming a rectangular area on the TZ plane, corresponding to a cuboid area in three-dimensional space.
[0038] Specifically, the avoidance zone is determined by two diagonal points, each containing boundary values in both the rotation axis direction and the elevation axis direction. For example... Figure 3In the illustrated embodiment, BlockPoint1 and BlockPoint2 are the first and second diagonal points of the avoidance interval Block1 corresponding to the loading interlock chamber LLB, respectively. BlockPoint1 is formed by the first rotational boundary value BlockPoint1.T in the rotational axis direction and the first lifting boundary value BlockPoint1.Z in the lifting axis direction of the avoidance interval Block1; BlockPoint2 is formed by the second rotational boundary value BlockPoint2.T in the rotational axis direction and the second lifting boundary value BlockPoint2.Z in the lifting axis direction of the avoidance interval Block1. BlockPoint3 and BlockPoint4 are the first and second diagonal points of the avoidance interval Block2 corresponding to the loading interlock chamber LLA, respectively. Similar to the definition of the avoidance zone Block 1 of the interlocking chamber LLB, BlockPoint 3 is jointly formed by the first rotational boundary value BlockPoint1.T in the rotational axis direction and the first lifting boundary value BlockPoint1.Z in the lifting axis direction of the avoidance zone Block 2; BlockPoint 4 is jointly formed by the second rotational boundary value BlockPoint2.T in the rotational axis direction and the second lifting boundary value BlockPoint2.Z in the lifting axis direction of the avoidance zone Block 2.
[0039] Combination Figures 5-7As shown, after the robotic arm models all avoidance zones based on its own coordinate system, it automatically determines whether the direct movement from the current position `cur` to the target position `aim` will pass through any avoidance zone before each motion task is executed. Specifically, the judgment logic is as follows: Let the rotation angle of the current position be `cur.T`, the height of the current position be `cur.Z`, the rotation angle of the target position be `aim.T`, and the height of the target position be `aim.Z`. If the angle range covered by the change in rotation angle from the current position's rotation angle `cur.T` to the target position's rotation angle `aim.T` intersects with the rotation angle range of the avoidance zone from `BlockPoint1.T` to `BlockPoint2.T`, and simultaneously, the height range covered by the change in height from the current position's height `cur.Z` to the target position's height `aim.Z` intersects with the height range of the avoidance zone from `BlockPoint1.Z` to `BlockPoint2.Z`, then it is determined that the direct movement will pass through the avoidance zone, i.e., there is interference, and avoidance planning should be initiated. Conversely, if there is only an intersection in one dimension and no intersection in the other dimension, it is determined that there is no interference. It's important to clarify that the angle range covered by the change from the current rotation angle cur.T to the target rotation angle aim.T refers to the actual range of motion, including scenarios where the robotic arm rotates infinitely across 0°. For example, for a robotic arm rotating infinitely, after moving 360°, it starts rotating from 0° again. That is, if the current rotation angle cur.T is 350° and the target rotation angle aim.T is 30°, the actual motion might be 350°~30°, i.e., a minor arc of 40°. For example, for a robotic arm that cannot rotate infinitely, if the allowable range of motion for the T-axis is -30°~270°, then moving from 250° to 10° would be a major arc of 240°.
[0040] This dual-dimensional joint judgment logic can accurately identify the interference risk of the robotic arm in actual three-dimensional space and avoid false triggering. When interference is detected in direct motion, the robotic arm will switch to an avoidance trajectory that passes through the safety line.
[0041] Furthermore, after modeling the avoidance zone, safety lines are designed. Considering the diverse interference situations in the actual workspace, this embodiment sets two types of safety lines, such as... Figure 8As shown in the diagram. The longitudinal safety line is a trajectory line where the T-axis is fixed at a specific angle, and the Z-axis moves freely within the teaching range Z1 to Z2. It is suitable for scenarios requiring height adjustments at a fixed azimuth angle to avoid interference zones. Z1 and Z2 are the upper and lower limits of the teaching range for the vertical axis Z in the longitudinal safety line, respectively. The transverse safety line is a trajectory line where the Z-axis is fixed at a specific height, and the T-axis moves freely within the teaching range T1 to T2. It is suitable for scenarios requiring large-scale rotations on a fixed height plane to avoid interference zones. T1 and T2 are the lower and upper limits of the teaching range for the horizontal axis T in the transverse safety line, respectively. The parameters of both the longitudinal and transverse safety lines are set by the user based on the geometry of the chamber and the workstation layout, thus ensuring the adaptability of this method to different chamber environments.
[0042] Furthermore, the final step involves path optimization. When multiple feasible avoidance trajectories exist between the same set of starting and target positions, the method provided by this invention automatically calculates the total travel distance of each trajectory and selects the trajectory with the shortest total travel distance for execution, thereby maintaining the optimal efficiency of the robotic arm's handling. For example... Figure 9 As shown, the first trajectory starts from the current position, moves to the first safety point sp1, and then reaches the target position via the second safety point sp2. The second trajectory starts from the current position, moves to the third safety point sp3, and then reaches the target position via the fourth safety point sp4. The first and second safety points sp1 and sp2 are on the same safety line, while the third and fourth safety points sp3 and sp4 are on a different safety line. When the safety lines involved in the two trajectories are parallel, by comparing the sum of the rotation distance from the current position to the first safety point sp1 and the rotation distance from the second safety point sp2 to the target position, and the sum of the rotation distance from the current position to the third safety point sp3 and the rotation distance from the fourth safety point sp4 to the target position, the trajectory with the shorter total distance can be selected as the optimal path. This significantly reduces the computational load and improves the computational efficiency of obstacle avoidance planning.
[0043] Combination Figure 10 As shown, based on the cluster-type equipment in the semiconductor manufacturing equipment of this embodiment, two avoidance zones, Block 1 and Block 2, are established corresponding to the loading interlock chambers LLA and LLB, respectively. In addition, an avoidance zone, Block 3, is established to prevent interference between the arm joint and LLA or LLB during the robot's rotation. The function of Block 3 is to limit the interference area between the robot's elbow and LLA or LLB when the robot cannot directly rotate to the direction of the process module PM2 at the height of LLA or LLB. Based on the above three avoidance zones, two safety lines are designed, with the longitudinal safety line denoted as Safety Line 1 and the transverse safety line denoted as Safety Line 2.
[0044] The actual obstacle avoidance route of the robot between each workstation is as follows: When moving from process module PM1 to LLA or LLB, from buffer workstation to LLA or LLB, or from LLA or LLB to PM1 or Buffer, the robot moves to its position along the longitudinal safety line 1. That is, the robot first moves from the starting point to one end of safety line 1, then moves along the safety line to the other end, and finally moves to the target position.
[0045] When moving from process module PM2 to LLA or LLB, or from LLA or LLB to PM2, the robot moves to its position along the transverse safety line 2. That is, the robot first moves from the starting point to one end of safety line 2, then moves along safety line 2 to the other end, and finally moves to the target position.
[0046] For the mutual movement between PM1, PM2 and Buffer, since there is no spatial interference between the workstations, the robot can move directly into position without the need to perform avoidance planning.
[0047] Through the above method, this embodiment realizes the autonomous obstacle avoidance movement of the cylindrical coordinate system robot in complex chamber spatial layout, without the need to pre-fix the motion trajectory, and can automatically select the optimal path, effectively improving the robot's handling efficiency and scene adaptability.
Claims
1. A method for obstacle avoidance movement of a robotic arm, characterized in that, Includes the following steps: At least one avoidance zone is established in the workspace of the robot arm, and the avoidance zone is jointly defined by the angular range in the direction of the rotation axis and the height range in the direction of the lifting axis. During the process of the robotic arm moving directly from its current position to the target position, it is determined whether there is an intersection between the angle range traversed by the rotation axis and the angle range of the avoidance zone, and whether there is an intersection between the height range traversed by the lifting axis and the height range of the avoidance zone. If both dimensions intersect, interference is determined, and an avoidance trajectory passing through at least one safety line is planned so that the robot arm can move around the avoidance zone to the target position.
2. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: The safety line includes a longitudinal safety line and / or a transverse safety line; the longitudinal safety line is a trajectory line with a fixed rotation axis angle and a variable lifting axis height; the transverse safety line is a trajectory line with a fixed lifting axis height and a variable rotation axis angle.
3. The obstacle avoidance motion method for a robotic arm according to claim 2, characterized in that: The height range of the lifting axis of the longitudinal safety line and the angle range of the rotation axis of the transverse safety line are set by the user according to the actual workspace.
4. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: If, during the process of the robotic arm moving directly from its current position to the target position, only the angle range traversed by the rotation axis intersects with the angle range of the avoidance zone, while the height range traversed by the lifting axis does not intersect with the height range of the avoidance zone, or only the height range traversed by the lifting axis intersects with the height range of the avoidance zone, while the angle range traversed by the rotation axis does not intersect with the angle range of the avoidance zone, then it is determined as non-interference.
5. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: The avoidance zone is a cuboid space region defined by two diagonal points, which correspond to two boundary values in the direction of rotation axis and two boundary values in the direction of elevation axis, respectively.
6. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: Also includes: When multiple feasible avoidance trajectories exist, calculate the total travel distance for each trajectory and select the trajectory with the shortest total travel distance to execute.
7. The obstacle avoidance motion method for a robotic arm according to claim 6, characterized in that: When multiple feasible avoidance trajectories involve parallel safety lines, the comparison of total travel distances simplifies to: Calculate the first sum and the second sum. The first sum is the sum of the rotation distance from the current point to the first safety point and the rotation distance from the second safety point to the target point. The second sum is the sum of the rotation distance from the current point to the third safety point and the rotation distance from the fourth safety point to the target point. Compare the first sum with the second sum, and select the trajectory with the smaller sum to execute.
8. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: The avoidance trajectory includes: moving from the current position to a first position on the safety line, moving along the safety line to a second position, and then moving from the second position to the target position.
9. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: The workspace contains multiple avoidance zones, each corresponding to a different spatial interference region.
10. The obstacle avoidance motion method for a robotic arm according to claim 1, characterized in that: The decision-making process is performed automatically before each motion task is executed by the robotic arm.