A CT imaging robot scan path collaborative planning method and system

Abstract

The embodiment relates to the technical field of robot path planning, and provides a CT image robot scanning path cooperative planning method and system. In view of the problem that single configuration path search leads to insufficient CT scanning arc angle, the application constructs and discretizes an arc scanning path, inversely kinematically solves each road point to obtain a plurality of configuration solutions, carries out collision and singularity detection after combination, screens a legal configuration set, determines a flag combination configuration according to the rules of preferentially keeping the same configuration and switching and replacing the configuration when necessary, and generates a multi-configuration splicing joint space path with the maximum arc angle by splicing continuous road points. By utilizing the inverse kinematics multi-solution and the non-overlapping characteristics of the reachable regions of different configurations, the scanning range is expanded through configuration switching, and the CT scanning arc angle and the imaging quality are significantly improved.

Description

Technical Field

[0001] This embodiment relates to the field of robot path planning technology, specifically to a collaborative planning method and system for CT imaging robot scanning paths. Background Technology

[0002] Computed tomography (CT) is an important diagnostic technique with advantages such as being non-contact and minimally invasive. It is an extremely important modality in the field of medical imaging and plays a crucial role in the diagnosis of head and neck diseases and chest and abdominal diseases.

[0003] Traditional O-arm CT scanners are the most widely used CT equipment in medical institutions. They employ a complete circular structure, enabling real-time 3D imaging during surgery to assist surgeons in precise positioning and manipulation, particularly suitable for spinal screw implantation. However, due to the structural limitations of O-arm CT scanners, they can only perform image reconstruction along a fixed trajectory, lacking flexibility. Furthermore, the O-arm structure significantly encroaches on the surgeon's operating space, requiring them to first locate the patient before performing the surgical procedure, making it difficult to synchronize surgical operations with 3D imaging. Additionally, O-arm CT scanners are bulky, occupying excessive operating room space and placing certain demands on spatial layout.

[0004] In recent years, C-arm CT scanners have been used more and more frequently, and the C-arm system has been well integrated with digital flat-panel detector cone-beam computed tomography (CBCT). The C-arm system can provide coverage of over 360 degrees and frees up more surgical space. However, the C-arm system has disadvantages such as being heavy and inconvenient to move. Furthermore, limited by the fixed gantry structure, the distance between the detector and the X-ray source in the C-arm is always fixed, making it unable to adapt to the flexible trajectory requirements of more complex scenarios.

[0005] Compared with the two structures mentioned above, the dual-robot system has advantages such as flexible scanning trajectory, easy storage, and relatively compact size, but it still has limitations in achieving a complete 360° circular arc scan.

[0006] In CT imaging, a sufficiently large circular scanning angle is a prerequisite for ensuring image quality, while existing image reconstruction algorithms typically require at least a 180° circular angle. In current technologies, robots struggle to plan large-angle circular paths under constraints such as task space, singularity, arm span, and collision avoidance, leading to insufficient image quality. Therefore, overcoming these limitations to enable robots to achieve sufficiently large circular angles is a key issue for the application of dual-robot systems in radiological imaging.

[0007] Currently, most robot path planning methods employ mainstream approaches, such as the Rapidly-exploring Random Trees (RRT) algorithm. These methods, based on a sampling approach, offer high planning efficiency, but their potential problem lies in their ability to solve paths within a single configuration within a limited timeframe. This limitation makes it difficult for robots to consider the coupling relationships between different configurations, hindering the planning of a path with the maximum arc angle and resulting in unsatisfactory imaging quality. As the demand for robots to perform complex tasks in radiographic imaging continues to grow, how to achieve collaborative path planning for maximizing arc angles under multiple constraints through flexible configuration switching has become a crucial technical problem that urgently needs to be solved. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this embodiment proposes a collaborative planning method and system for CT imaging robot scanning paths. This embodiment innovatively introduces configuration switching into path planning as a means of avoiding collision constraints. Through flexible configuration selection, the path planning problem under task space constraints is transformed into the acquisition and combination of inverse kinematics, thereby simplifying the path planning process and achieving optimal arc angles under multiple constraints. This enables dual-robot collaborative path planning for maximizing arc angles under the aforementioned constraints.

[0009] In a first aspect, this application provides a collaborative planning method for scanning paths of CT imaging robots, applicable to a dual-robot CT imaging system consisting of a left robot and a right robot, comprising: In Cartesian space, the desired circular arc scanning path is constructed according to the requirements of the CT scan task. The circular arc scanning path is a circular arc path around the object being measured. The circular arc scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose. For each discrete waypoint, inverse kinematics solutions are performed for both the left and right robots to obtain all configuration solutions for that waypoint. At each waypoint, the configuration solutions of the left robot and the right robot are combined, and collision detection and singularity detection are performed on the combined configuration to select the set of legal configurations that meet the collision avoidance constraint and singularity avoidance constraint. Based on the set of legal configurations, the flag combination configuration of each waypoint is determined according to the flag combination configuration selection rule that prioritizes maintaining configuration consistency between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained. Search for all continuous waypoint sets with existing sign combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration splicing joint spatial scan path with the largest arc angle.

[0010] Furthermore, the inverse kinematics solution is based on the Pieper criterion and employs an analytical method to obtain all joint configuration solutions for the left and right robots at each path point.

[0011] Furthermore, the end-effector pose is determined based on the coordinates of the arc center, the arc length of the corresponding robot on the arc path, and the angles of each path point, and the left robot and the right robot have a one-to-one correspondence at each path point based on geometric constraints.

[0012] Furthermore, the collision detection includes collision detection between the robot and the environment, collision detection between the left robot and the right robot, and collision detection between the robot and its own body; the singularity detection is based on the determination of the minimum singular value of the Jacobian matrix.

[0013] Furthermore, the selection rules for the symbol combination configuration include: If the set of valid configurations for the current waypoint is not empty and the previous waypoint has a flag combination configuration, then the configuration that belongs to the same type as the previous waypoint's flag combination configuration is selected as the current waypoint's flag combination configuration. If there is no combination configuration in the current set of legal configurations that belongs to the same configuration as the previous waypoint, then the combination configuration that is farthest from the colliding object in the set of legal configurations is selected as the flag combination configuration of the current waypoint, and a configuration switch is determined. If the set of legal configurations for the current waypoint is not empty and the previous waypoint does not have a flag combination configuration, then the combination configuration farthest from the colliding object in the set of legal configurations is selected as the flag combination configuration for the current waypoint.

[0014] Furthermore, the search for the set of all continuous waypoints with existing flag combinations includes: Starting from the beginning point of the circular path, traverse all waypoints and determine whether each waypoint has a combination of markers. Group consecutive waypoints with a combination of sign configurations into a single set of consecutive waypoints. The arc angles corresponding to each set of continuous waypoints are sorted, and the set of continuous waypoints with the largest arc angle is selected as the spatial scanning path of the multi-configuration splicing joint.

[0015] Furthermore, the configuration switching involves switching the robot from its current joint configuration to another joint configuration corresponding to the same inverse kinematic solution under the same end-effector pose while keeping the end-effector pose unchanged. This utilizes the non-overlapping characteristics of the reachable regions of different configurations to expand the overall scanning range, thereby avoiding collisions or singularities and increasing the scanning arc angle.

[0016] Furthermore, the left robot and the right robot are both six-degree-of-freedom collaborative robots, and the same end-effector pose corresponds to multiple sets of different joint configuration solutions.

[0017] Secondly, this application also proposes a CT imaging robot scanning path collaborative planning system, applicable to a dual-robot CT imaging system consisting of a left robot and a right robot, characterized by comprising: The path discretization module is used to construct the desired circular arc scanning path in Cartesian space according to the requirements of the CT scan task. The circular arc scanning path is a circular arc path around the object being measured. The circular arc scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose. The inverse kinematics solution module is used to perform inverse kinematics solutions for the left and right robots for each discrete waypoint, and obtain all configuration solutions for that waypoint. The configuration filtering module is used to combine the configuration solutions of the left robot and the right robot at each waypoint, and to perform collision detection and singularity detection on the combined configurations to filter out the set of legal configurations that meet the collision avoidance constraints and singularity avoidance constraints. The sign configuration selection module is used to determine the sign combination configuration of each waypoint based on the set of legal configurations, according to the sign combination configuration selection rule that prioritizes maintaining the same configuration between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained. The path stitching module is used to search for all continuous waypoint sets with existing marker combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration stitching joint space scan path with the largest arc angle.

[0018] Furthermore, the path discretization module, inverse solution module, configuration filtering module, marker configuration selection module, and path splicing module are configured to execute the steps of the CT imaging robot scanning path collaborative planning method as described in the first aspect.

[0019] Compared with the prior art, the beneficial effects of this embodiment are as follows: This embodiment proposes a collaborative planning method for CT imaging robot scanning paths. It innovatively introduces configuration switching into path planning as a means to avoid collision constraints. Through flexible configuration selection, the path planning problem under task space constraints is transformed into the acquisition and combination of inverse kinematics, thereby simplifying the path planning process and achieving the optimality of the arc angle under multiple constraints. This realizes the dual-robot collaborative path planning for maximizing the arc angle under the aforementioned constraints. By implementing configuration switching at constraint boundaries, the method in this embodiment can maximize the arc path angle in a fixed deployment position of the two robots, thereby effectively ensuring the imaging quality of radiographic imaging. Simultaneously, in environments with more stringent collision avoidance constraints, such as confined spaces, this method, through flexible configuration selection and frequent configuration switching, ensures sufficient angles for imaging, significantly improving the applicability of the dual-robot system for performing radiographic tasks in complex and confined spaces. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of this embodiment, the accompanying drawings used in the embodiment will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this embodiment and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the steps of the CT imaging robot scanning path collaborative planning method disclosed in this embodiment.

[0022] Figure 2 This is a schematic diagram illustrating the simulated scanning process of a CT imaging robot along a 90° circular arc path using traditional path planning methods; where... Figure 2 (a) shows the robotic arm configuration when the scanning arc angle is 7.5 degrees in the simulation environment. Figure 2 (b) shows the configuration of the robotic arm when the scanning arc angle is 97.5 degrees in the simulation environment.

[0023] Figure 3 This is a schematic diagram illustrating the simulated scanning process of a 194° circular arc path using the method proposed in this embodiment for a CT imaging robot; wherein... Figure 3 (a) shows the robotic arm configuration when the scanning arc angle is 7.5 degrees in the simulation environment. Figure 3 (b) shows the robotic arm configuration when the scanning arc angle is 110 degrees in the simulation environment; Figure 3 Image (c) shows the robotic arm configuration when the scanning arc angle is 110 degrees in the simulation environment. Figure 3 The configuration of the robotic arm in the simulation environment is shown in (d) when the scanning arc angle is 201.5 degrees.

[0024] Figure 4 This is a schematic diagram illustrating the actual scanning process of the 194° circular arc path using the method proposed in this embodiment of the CT imaging robot; wherein... Figure 4 (a) shows the configuration of the robotic arm when the scanning arc angle is 7.5 degrees under physical experimental conditions. Figure 4 (b) shows the configuration of the robotic arm when the scanning arc angle is 110 degrees under physical experimental conditions; Figure 4Image (c) shows the robotic arm configuration when the scanning arc angle is 110 degrees under physical experimental conditions. Figure 4 The configuration of the robotic arm in the middle (d) is when the scanning arc angle is 201.5 degrees under physical experimental conditions.

[0025] Figure 5 These are comparison images of CT scans performed by a CT imaging robot along circular arc paths of 90° and 194°; among them, Figure 5 (a) is a schematic diagram of the cylindrical mold; Figure 5 (b) is a schematic diagram of the customized school emblem prototype; Figure 5 Image (c) shows the imaging results of a cylindrical phantom under a 90° CT arc planned using the traditional path planning method. Figure 5 Image (d) shows the imaging result of the school emblem phantom under a 90° CT arc planned using the traditional path planning method. Figure 5 (e) is the imaging result of a cylindrical phantom under a 194° CT arc planned by the method proposed in this invention; Figure 5 (f) shows the imaging result of the school emblem phantom under a 194° CT arc as planned by the method proposed in this invention.

[0026] Figure 6 This is a schematic diagram of the CT imaging robot scanning path collaborative planning system disclosed in this embodiment. Detailed Implementation

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which these embodiments belong; the terminology used herein and in the specification of the application is for the purpose of describing particular embodiments only and is not intended to limit these embodiments; the terms "comprising" and "having," and any variations thereof, in the specification of these embodiments and the foregoing drawings, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification of these embodiments and the foregoing drawings are used to distinguish different objects, not to describe a particular order.

[0028] The implementation details of the technical solution in this embodiment are described in detail below: Firstly, this embodiment proposes a collaborative planning method for CT imaging robot scanning paths, applicable to dual-robot CT imaging systems consisting of a left robot and a right robot. For example... Figure 1 As shown, the method includes: S101, In Cartesian space, construct the desired circular arc scanning path according to the requirements of the CT scan task. The circular arc scanning path is a circular arc path around the object being measured. Discretize the circular arc scanning path according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose.

[0029] Furthermore, the end-effector pose is determined based on the coordinates of the arc center, the arc length of the corresponding robot on the arc path, and the angles of each path point, and the left robot and the right robot have a one-to-one correspondence at each path point based on geometric constraints.

[0030] Specifically, in this embodiment, the circular scanning path is an arc path surrounding the object being measured, determined by the coordinates of the arc center, the arc length of the left robot on the arc path, the arc length of the right robot on the arc path, and the start and end angles of the arc path. The circular scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints, each waypoint corresponding to an end pose. In this embodiment, the preset angle step size is 0.5°, and the circular path covering a 360° range is discretized, resulting in a total of 721 discrete waypoints.

[0031] The left and right robots have a one-to-one correspondence at each waypoint based on geometric constraints: the left and right robots use the same arc center as a reference and synchronously sample according to the same discrete angle increment. The waypoints of the left and right robots appear in pairs at the same angle position, thus establishing a one-to-one correspondence between the waypoints of the left and right robots.

[0032] The desired pose of the left robot at each waypoint is determined by the following formula:

[0033] in, Indicates the starting calculated angle of the arc path. This represents the angular increment of the discretized sampling. Indicates the coordinate position of the center of the arc; This indicates the arc length of the left robot along the circular path; Indicates the left robot is The corresponding arc coordinate position, Indicates the left robot is The corresponding arc posture.

[0034] The desired pose of the discrete waypoints of the right robot is as follows:

[0035] in, Indicates the starting calculated angle of the arc path. This represents the angular increment of the discretized sampling. Indicates the coordinate position of the center of the arc; This indicates the arc length of the right robot along the circular path; Indicates the right robot is The corresponding arc coordinate position, Indicates the right robot is The corresponding arc posture.

[0036] Through the aforementioned geometric constraints, the left and right robots form a definite pose correspondence at each discrete waypoint, ensuring that subsequent inverse kinematics solutions and configuration combinations can be performed synchronously at paired waypoints.

[0037] S102, for each discrete waypoint, perform inverse kinematics solutions for the left and right robots respectively to obtain all configuration solutions for that waypoint.

[0038] Furthermore, the inverse kinematics solution is based on the Pieper criterion and employs an analytical method to obtain all joint configuration solutions for the left and right robots at each path point.

[0039] Specifically, since the UR20 collaborative robot used in this embodiment satisfies the Pieper criterion, its inverse kinematics has an analytical solution. Therefore, all configuration solutions at this waypoint can be obtained directly using analytical methods without iterative approximation, resulting in high computational efficiency and ensuring solution completeness. The Pieper criterion is applicable to six-DOF robot configurations where the last three joint axes intersect at a single point or are parallel to each other. When this criterion is satisfied, the inverse kinematics problem can be decoupled into two subproblems: inverse position solution and inverse attitude solution. The joint angles are calculated sequentially using closed-form analytical formulas, thereby obtaining all joint configuration solutions at this end-effector pose in one go, avoiding the problem of numerical iteration potentially getting trapped in local optima or missing feasible solutions.

[0040] In this embodiment, the left robot exists at the current waypoint. The right robot has a configuration solution at the current waypoint. Configurational solutions. Defined separately: Left robot configuration solution: ; Solution to the configuration of the right robot: ; Each configuration solution corresponds to a complete set of joint angle configurations, representing a reachable posture of the robot at that waypoint. All configuration solutions obtained analytically constitute a complete candidate space for subsequent configuration combinations and selection, ensuring that no feasible joint configurations are overlooked, and providing a data foundation for subsequent switching between multiple configurations.

[0041] S103. At each waypoint, combine the configuration solutions of the left robot and the right robot, and perform collision detection and singularity detection on the combined configuration to select a set of legal configurations that meet the collision avoidance constraint and singularity avoidance constraint.

[0042] Furthermore, the collision detection includes collision detection between the robot and the environment, collision detection between the left robot and the right robot, and collision detection between the robot and its own body; the singularity detection is based on the determination of the minimum singular value of the Jacobian matrix.

[0043] Specifically, in this embodiment, at each discrete waypoint, all configuration solutions of the left and right robots are combined to obtain... Combinatorial configuration. Noted as:

[0044] For each combination configuration Each includes one joint configuration of the left robot and one joint configuration of the right robot, representing a cooperative posture of the two robots at that waypoint. Through the analysis of all... By performing a comprehensive test on the combined configurations, all possible cooperative configuration schemes at the waypoint can be exhausted, avoiding the omission of potential feasible solutions.

[0045] Collision detection is performed on each combined configuration. This collision detection includes collision detection between the robot and the environment, between the left and right robots, and between the robot and its own body. Specifically, collision detection between the robot and the environment determines whether the robot links or end effectors geometrically interfere with external obstacles; collision detection between the left and right robots determines whether the two robots interfere with each other during cooperative movement; and collision detection between the robot and its own body determines whether there are self-collisions between the robot links. If no collision is detected in any of the above three types of collision detection, the combined configuration is determined to satisfy the collision avoidance constraints.

[0046] Singularity detection is performed for each configuration. The singularity detection is based on the minimum singular value of the Jacobian matrix. Specifically, the Jacobian matrices of the left and right robots under the current joint configuration are calculated, and their minimum singular values ​​are determined. If the minimum singular value is lower than a preset singularity threshold, the robot is determined to be near a singular configuration, and the configuration does not satisfy the singularity avoidance constraint. If the minimum singular values ​​of both the left and right robots are greater than or equal to the preset singularity threshold, the configuration is determined to be non-singular and satisfies the singularity avoidance constraint.

[0047] If a composite configuration is neither colliding nor singular, then record it in the set of legal configurations for that waypoint. Through the above combination and screening process, each waypoint can obtain a set of legal configurations containing all feasible cooperative solutions, providing a candidate space for the selection of subsequent marker configurations.

[0048] S104. Based on the set of legal configurations, determine the flag combination configuration of each waypoint according to the flag combination configuration selection rule that prioritizes maintaining configuration consistency between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained.

[0049] Specifically, to ensure path continuity and improve CT scan efficiency, a unique marker combination configuration needs to be determined from the set of legal configurations at each waypoint, serving as the representative configuration for the cooperative motion of the two robots at that waypoint. The marker combination configurations are stored in array form for each waypoint, and a sequence of marker combination configurations is established sequentially along the arc path direction, providing a continuous index for subsequent path splicing.

[0050] The rule for selecting the configuration combination of the markers follows these steps: First, determine whether the set of legal configurations for the current waypoint is not empty. If the set of legal configurations is not empty, it indicates that the current waypoint has a feasible configuration, and the current waypoint is determined to be reachable. Then, further processing is performed based on the state of the previous waypoint: (1) If the set of legal configurations of the current waypoint is not empty and the previous waypoint has a sign combination configuration, then the combination configuration that belongs to the same type as the previous waypoint's sign combination configuration shall be selected as the current waypoint's sign combination configuration.

[0051] Specifically, belonging to the same configuration means that the joint configuration patterns of the left and right robots correspond to the same configuration category, which can be determined by the symbol pattern of each joint angle, elbow orientation, or wrist posture category. For example, for a six-DOF robot that satisfies the Pieper criterion, all configuration solutions can be divided into several configuration categories according to shoulder posture, elbow orientation, and wrist rotation state. If the current combined configuration and the previous waypoint marker combined configuration both belong to the same category on the left and right robots, then they are determined to belong to the same configuration. If such a combined configuration exists, it is preferentially selected as the marker combined configuration of the current waypoint, thereby maintaining the configuration consistency between adjacent waypoints and reducing the motion impact and control complexity caused by unnecessary configuration switching.

[0052] (2) If there is no combination configuration in the current set of legal configurations that belongs to the same configuration as the previous waypoint, then select the combination configuration that is farthest from the colliding object in the set of legal configurations as the flag combination configuration of the current waypoint, and determine that a configuration switch has occurred; Specifically, if there is no combination configuration of the same type as the previous waypoint in the current set of legal configurations, it indicates that the original configuration of the previous waypoint cannot be maintained at the current waypoint. In this case, the minimum Euclidean distance between the robot link envelope corresponding to each combination configuration in the current set of legal configurations and the colliding object in the environment is calculated. The combination configuration with the largest minimum distance, i.e., the furthest from the colliding object, is selected as the flag combination configuration of the current waypoint. Compared to the previous waypoint, the flag combination configuration has undergone a configuration switch, changing from the previous configuration to the current configuration. This configuration switch operation fully utilizes the multi-solution characteristic of robot inverse kinematics, switching the two robots from a configuration prone to collision or singularities to a safer configuration while maintaining the end-effector pose, thereby expanding the effective scanning range.

[0053] (3) If the set of legal configurations of the current waypoint is not empty and there is no sign combination configuration of the previous waypoint, then the combination configuration that is farthest from the colliding object in the set of legal configurations is selected as the sign combination configuration of the current waypoint.

[0054] Specifically, in this embodiment, if the set of legal configurations for the current waypoint is not empty and the previous waypoint does not have a flag combination configuration, then the current waypoint is at the beginning of a continuous path. At this time, the minimum distance between each combination configuration in the current set of legal configurations and the colliding object is calculated, and the combination configuration farthest from the colliding object is selected as the flag combination configuration of the current waypoint to ensure that the safest configuration configuration is used at the beginning of the path.

[0055] Furthermore, if the set of valid configurations for the current waypoint is empty, that is, there is no valid configuration, then the current waypoint is determined to be unreachable, and no flag combination configuration is generated for the waypoint, and the corresponding position is stored as an empty value.

[0056] By using the above-mentioned sign combination configuration selection rules, configuration consistency can be maintained between adjacent waypoints, and configuration switching can be performed only when necessary. This achieves path continuity and scanning efficiency while satisfying collision avoidance and oddity avoidance constraints.

[0057] S105, search for all continuous waypoint sets with existing marker combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration splicing joint spatial scan path with the largest arc angle.

[0058] In this embodiment, all discrete waypoints are traversed, and continuous waypoints with combined marker configurations are spliced ​​together to form a joint space path with multiple configurations.

[0059] The search for the set of all continuous waypoints with existing flag combinations includes: Starting from the initial waypoint of the circular path, traverse all waypoints and determine whether each waypoint has a marker combination configuration. Specifically, sequentially read the marker combination configuration data stored at each waypoint. If a valid marker combination configuration exists at the waypoint, then the waypoint is determined to be a candidate waypoint for a certain continuous path; if the marker combination configuration at the waypoint is empty, then the waypoint is determined to be unreachable and does not belong to any continuous path.

[0060] Consecutive waypoints with a marker combination configuration are grouped into a continuous waypoint set. During traversal, the reachability of the current waypoint and its preceding waypoint is jointly determined: if both the current waypoint and the preceding waypoint have a marker combination configuration, the current waypoint is within a continuous path; if the current waypoint has a marker combination configuration but the preceding waypoint does not, the current waypoint is the starting waypoint of the continuous path; if the current waypoint does not have a marker combination configuration but the preceding waypoint does, the preceding waypoint is the ending point of the continuous path, and a continuous waypoint set is recorded; if neither the current waypoint nor the preceding waypoint has a marker combination configuration, neither waypoint belongs to any continuous path. Through the above determination, all discrete waypoints can be divided into several independent continuous waypoint sets, where the waypoint indices in each set are consecutive and all have a marker combination configuration.

[0061] The arc angles corresponding to each set of continuous waypoints are sorted, and the set of continuous waypoints with the largest arc angle is selected as the spatial scanning path for the multi-configuration splicing joint. Specifically, the arc angle range covered by each set of continuous waypoints is calculated, that is, the angle span between the starting and ending waypoints within the set. The arc angles of all sets of continuous waypoints are compared and sorted, and the set of continuous waypoints containing the most waypoints and the largest arc angle is selected. The maximum arc motion angle achievable by the dual-robot system is determined by the path range corresponding to this set. Simultaneously, since each waypoint in the set of continuous waypoints records a corresponding combination of marker configurations, a discrete joint motion path corresponding to the maximum arc angle is obtained.

[0062] Furthermore, the configuration switching in this embodiment is achieved based on the multi-solution characteristic of robot inverse kinematics. Under the same end-effector pose, the robot can correspond to multiple different joint configurations. The essence of configuration switching is to switch the robot from a configuration prone to collisions or singularities to a more optimal configuration while keeping the end-effector pose unchanged, thereby expanding the workspace. This can be compared to a human hand grasping an object, using different postures (such as forward grasping, reverse grasping, or side grasping) and selecting a more suitable posture in different spatial positions to complete a wider range of operations. Through the above-mentioned multi-configuration stitching and configuration switching strategy, while satisfying collision avoidance constraints and singularity constraints, the scanning path range is effectively expanded, thereby improving the CT scan arc angle.

[0063] Furthermore, the configuration switching involves switching the robot from its current joint configuration to another joint configuration corresponding to the same inverse kinematic solution under the same end-effector pose while keeping the end-effector pose unchanged. This utilizes the non-overlapping characteristics of the reachable regions of different configurations to expand the overall scanning range, thereby avoiding collisions or singularities and increasing the scanning arc angle.

[0064] Specifically, in a dual-robot CT imaging system, the left and right robots each carry an X-ray emission tube and a detector, and their relative poses directly determine the geometric relationship of the CT scan. When a robot moves along an arc path to a certain area, if the link posture under the current configuration results in the distance between the robot and environmental obstacles or another robot being less than a safe threshold, it is determined that the current configuration is restricted by collision constraints and cannot continue to extend the scanning path. At this time, based on the multiple solutions characteristic of robot inverse kinematics, while keeping the end effector pose completely unchanged, the current joint configuration is switched to the joint configuration corresponding to another set of inverse kinematic solutions under the same end effector pose. Since the spatial postures of each link of the robot are significantly different under different joint configurations, their corresponding reachable areas are also different, and even exhibit some non-overlapping characteristics; some spatial areas that are blocked by obstacles due to link interference in the current configuration may become reachable areas in another configuration. By selecting different configurations at different waypoints, the dual-robot system can utilize the non-overlapping nature of the reachable areas of different configurations to stitch together a continuous scanning path covering a larger arc range, thereby effectively avoiding collisions or singular constraints and significantly expanding the scanning arc angle. This can be compared to a human hand grasping an object, which can adopt different postures (such as forward grasping, reverse grasping, or side grasping) and select a more suitable posture at different spatial positions to complete a wider range of operations.

[0065] For example, in the experimental setup of this embodiment, when the dual-robot system uses MoveIt for traditional single-configuration path planning, the configuration is not optimal because it does not follow the rule of "the configuration furthest from the collision object" at the path starting point of 7.5°. Instead, it directly uses a default or randomly determined configuration. As the scanning angle increases, the robot links gradually approach the obstacle. When the scanning angle of the left robotic arm reaches 97.5°, its first and second links are about to collide with the ground. The system cannot continue to expand the scanning angle under the current configuration, and the scanning arc range only covers about 90° (from 7.5° to 97.5°).

[0066] Furthermore, the left robot and the right robot are both six-degree-of-freedom collaborative robots, and the same end-effector pose corresponds to multiple sets of different joint configuration solutions.

[0067] In this embodiment, both the left and right robots are UR20 collaborative robots, which are six-degree-of-freedom serial robots that satisfy the Pieper criterion. The last three joint axes intersect at a single point, giving the inverse kinematics problem a closed-form analytical solution. For a six-degree-of-freedom robot, given the end effector pose, solving the inverse kinematics problem typically involves decoupling the inverse position and inverse attitude solutions. Through trigonometric transformations and algebraic elimination, up to eight different joint configuration solutions can be obtained under the same end effector pose. These configuration solutions are distributed differently in joint space, corresponding to different combinations of shoulder pose, elbow orientation, and wrist rotation, i.e., different joint configurations.

[0068] For example, for a discrete waypoint on a circular path, the end-effector pose corresponding to this waypoint is uniquely determined by the coordinates of the arc center, the robot's arc length, and the current waypoint angle. When performing inverse kinematics on the left robot at this waypoint, the analytical method based on the Pieper criterion first calculates the wrist center position, then calculates the first three joint angles through geometric relationships, and finally calculates the last three joint angles based on the wrist posture. After a complete solution, eight sets of candidate joint configuration solutions are obtained. In actual experiments, all eight sets of candidate joint configuration solutions satisfy the motion range constraints of each robot joint and do not need to be eliminated. In the subsequent combination stage, these eight sets of left robot configuration solutions are combined one by one with the eight sets of right robot configuration solutions at the corresponding waypoints to form sixty-four combined configurations. Then, collision detection and singularity detection are performed on each of these sixty-four combined configurations, and the combined configurations that satisfy the collision avoidance constraint and singularity avoidance constraint are recorded in the legal configuration set of this waypoint. Finally, when splicing the joint paths, the combined configurations are spliced ​​according to the selection rules of the marked combined configurations from this legal configuration set. Since the same end-effector pose corresponds to multiple sets of different joint configuration solutions, the robot can flexibly select any one of these configurations based on spatial constraints when performing a task. This provides a solid theoretical foundation and a prerequisite for the configuration switching mechanism. Based on this multi-solution characteristic, this embodiment can adapt to complex spatial constraints and achieve large-area CT scanning by switching to different joint configuration solutions without changing the end-effector pose.

[0069] To verify the effectiveness of the CT imaging robot scanning path collaborative planning method proposed in this embodiment, the scanning arc angle under different path planning methods is compared and analyzed with reference to the accompanying drawings.

[0070] like Figure 2 The diagram shows a simulated scanning process of a CT imaging robot using a 90° circular arc path under traditional path planning methods. Figure 2 (a) shows the robotic arm configuration when the scanning arc angle is 7.5 degrees in the simulation environment. Figure 2 (b) shows the configuration of the robotic arm when the scanning arc angle is 97.5 degrees in the simulation environment.

[0071] Figure 3 This is a schematic diagram illustrating the simulated scanning process of a 194° circular arc path using the method proposed in this embodiment for a CT imaging robot. It should be noted that the initial configuration selection in the method proposed in this embodiment differs from the initial configuration selected in traditional path planning methods. Figure 3 (a) shows the robotic arm configuration when the scanning arc angle is 7.5 degrees in the simulation environment. Figure 3 (b) shows the robotic arm configuration when the scanning arc angle is 110 degrees in the simulation environment; Figure 3 Image (c) shows the robotic arm configuration when the scanning arc angle is 110 degrees in the simulation environment. Figure 3 (d) is the robotic arm configuration when the scanning arc angle is 201.5 degrees in the simulation environment; Figure 4 This is a schematic diagram illustrating the actual scanning process of the 194° circular arc path using the method proposed in this embodiment of the CT imaging robot; wherein... Figure 4 (a) shows the configuration of the robotic arm when the scanning arc angle is 7.5 degrees under physical experimental conditions. Figure 4 (b) shows the configuration of the robotic arm when the scanning arc angle is 110 degrees under physical experimental conditions; Figure 4 Image (c) shows the robotic arm configuration when the scanning arc angle is 110 degrees under physical experimental conditions. Figure 4 The configuration of the robotic arm in the middle (d) is when the scanning arc angle is 201.5 degrees under physical experimental conditions.

[0072] Figure 5 These are comparison images of CT scans performed by a CT imaging robot along circular arc paths of 90° and 194°. Figure 5 (a) is a schematic diagram of the cylindrical mold; Figure 5 (b) is a schematic diagram of the customized school emblem prototype; Figure 5 Image (c) shows the imaging results of a cylindrical phantom under a 90° CT arc planned using the traditional path planning method. Figure 5 Image (d) shows the imaging result of the school emblem phantom under a 90° CT arc planned using the traditional path planning method. Figure 5 (e) is the imaging result of a cylindrical phantom under a 194° CT arc planned by the method proposed in this invention; Figure 5 (f) shows the imaging result of the school emblem phantom under a 194° CT arc as planned by the method proposed in this invention.

[0073] In this experimental setup, when using MoveIt for path planning, the dual-robot system can only cover a scanning arc range of approximately 90°. Specifically, the dual robots move from 7.5° to 97.5°, and the path planning results are as follows: Figure 2As shown, when the scanning angle of the left robotic arm reaches 97.5°, its first and second links are about to collide with the ground, preventing the system from continuing to expand the scanning angle in the current configuration.

[0074] Ideally, a dual-arm system deployed in the same space can increase the scanning arc angle to 194 degrees by switching the arm configuration. To address this issue, the path planning method based on configuration switching proposed in this invention fully utilizes the non-overlapping characteristics of the workspaces of the arms under different configurations, achieving continuous path connection through configuration switching, thereby effectively expanding the scanning range. Experimental results show that after adopting the method of this invention, the scanning arc range is significantly increased from 90° to 194°. The path planning results and key configuration change process are as follows: Figure 3 As shown. Specifically, the dual robots move from 7.5° to 110° in configuration; when they reach 110°, they perform a configuration switch operation, and then continue to move from 110° to 201.5° in configuration, thereby achieving a large-scale continuous scan.

[0075] Secondly, this embodiment also proposes a CT imaging robot scanning path collaborative planning system, applicable to a dual-robot CT imaging system consisting of a left robot and a right robot. For example... Figure 6 As shown, the CT imaging robot scanning path collaborative planning system includes: The path discretization module 601 is used to construct the desired circular arc scanning path in Cartesian space according to the requirements of the CT scanning task. The circular arc scanning path is a circular arc path around the object being measured. The circular arc scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose. The inverse kinematics solution module 602 is used to perform inverse kinematics solutions for the left and right robots for each discrete waypoint, and obtain all configuration solutions for that waypoint. The configuration filtering module 603 is used to combine the configuration solutions of the left robot and the right robot at each waypoint, and to perform collision detection and singularity detection on the combined configurations to filter out the set of legal configurations that meet the collision avoidance constraints and singularity avoidance constraints. The sign configuration selection module 604 is used to determine the sign combination configuration of each waypoint based on the set of legal configurations, according to the sign combination configuration selection rule that prioritizes maintaining the same configuration between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained. The path stitching module 605 is used to search for all continuous waypoint sets with existing marker combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration stitching joint space scan path with the largest arc angle.

[0076] Furthermore, the path discretization module 601, inverse solution module 602, configuration screening module 603, marker configuration selection module 604, and path splicing module 605 are configured to execute the steps of the CT imaging robot scanning path collaborative planning method as described in the first aspect, which will not be repeated here.

[0077] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A collaborative planning method for scanning paths of CT imaging robots, applicable to a dual-robot CT imaging system consisting of a left robot and a right robot, characterized in that, include: In Cartesian space, the desired circular arc scanning path is constructed according to the requirements of the CT scan task. The circular arc scanning path is a circular arc path around the object being measured. The circular arc scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose. For each discrete waypoint, inverse kinematics solutions are performed for both the left and right robots to obtain all configuration solutions for that waypoint. At each waypoint, the configuration solutions of the left robot and the right robot are combined, and collision detection and singularity detection are performed on the combined configuration to select the set of legal configurations that meet the collision avoidance constraint and singularity avoidance constraint. Based on the set of legal configurations, the flag combination configuration of each waypoint is determined according to the flag combination configuration selection rule that prioritizes maintaining configuration consistency between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained. Search for all continuous waypoint sets with existing sign combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration splicing joint spatial scan path with the largest arc angle.

2. The CT imaging robot scanning path collaborative planning method according to claim 1, characterized in that, The inverse kinematics solution is based on the Pieper criterion and uses an analytical method to obtain all joint configuration solutions for the left and right robots at each path point.

3. The CT imaging robot scanning path collaborative planning method according to claim 2, characterized in that, The end-effector pose is determined based on the coordinates of the arc center, the arc length of the corresponding robot on the arc path, and the angles of each path point. Furthermore, the left and right robots have a one-to-one correspondence at each path point based on geometric constraints.

4. The CT imaging robot scanning path collaborative planning method according to claim 3, characterized in that, The collision detection includes collision detection between the robot and the environment, collision detection between the left robot and the right robot, and collision detection between the robot and its own body; the singularity detection is based on the determination of the minimum singular value of the Jacobian matrix.

5. The CT imaging robot scanning path collaborative planning method according to claim 1, characterized in that, The rules for selecting the logo combination configuration include: If the set of valid configurations for the current waypoint is not empty and the previous waypoint has a flag combination configuration, then the configuration that belongs to the same type as the previous waypoint's flag combination configuration is selected as the current waypoint's flag combination configuration. If there is no combination configuration in the current set of legal configurations that belongs to the same configuration as the previous waypoint, then the combination configuration that is farthest from the colliding object in the set of legal configurations is selected as the flag combination configuration of the current waypoint, and a configuration switch is determined. If the set of legal configurations for the current waypoint is not empty and the previous waypoint does not have a flag combination configuration, then the combination configuration farthest from the colliding object in the set of legal configurations is selected as the flag combination configuration for the current waypoint.

6. The CT imaging robot scanning path collaborative planning method according to claim 5, characterized in that, The search for the set of all continuous waypoints with existing flag combinations includes: Starting from the beginning point of the circular path, traverse all waypoints and determine whether each waypoint has a combination of markers. Group consecutive waypoints with a combination of sign configurations into a single set of consecutive waypoints. The arc angles corresponding to each set of continuous waypoints are sorted, and the set of continuous waypoints with the largest arc angle is selected as the spatial scanning path of the multi-configuration splicing joint.

7. The CT imaging robot scanning path collaborative planning method according to claim 6, characterized in that, The configuration switching involves switching the robot from its current joint configuration to another joint configuration corresponding to the same inverse kinematic solution while keeping the end-effector pose unchanged. This utilizes the non-overlapping characteristics of the reachable regions of different configurations to expand the overall scanning range, thereby avoiding collisions or singularities and increasing the scanning arc angle.

8. The CT imaging robot scanning path collaborative planning method according to claim 1, characterized in that, The left robot and the right robot are both six-degree-of-freedom collaborative robots, and the same end-effector pose corresponds to multiple sets of different joint configuration solutions.

9. A CT imaging robot scanning path collaborative planning system, applicable to a dual-robot CT imaging system consisting of a left robot and a right robot, characterized in that, include: The path discretization module is used to construct the desired circular arc scanning path in Cartesian space according to the requirements of the CT scan task. The circular arc scanning path is a circular arc path around the object being measured. The circular arc scanning path is discretized according to a preset angle step size to obtain a number of discrete waypoints. Each waypoint corresponds to an end pose. The inverse kinematics solution module is used to perform inverse kinematics solutions for the left and right robots for each discrete waypoint, and obtain all configuration solutions for that waypoint. The configuration filtering module is used to combine the configuration solutions of the left robot and the right robot at each waypoint, and to perform collision detection and singularity detection on the combined configurations to filter out the set of legal configurations that meet the collision avoidance constraints and singularity avoidance constraints. The sign configuration selection module is used to determine the sign combination configuration of each waypoint based on the set of legal configurations, according to the sign combination configuration selection rule that prioritizes maintaining the same configuration between adjacent waypoints and switches to an alternative configuration that satisfies the constraints when the same configuration cannot be maintained. The path stitching module is used to search for all continuous waypoint sets with existing marker combination configurations, select the continuous waypoint set with the largest arc angle, and generate a multi-configuration stitching joint space scan path with the largest arc angle.

10. The CT imaging robot scanning path collaborative planning system according to claim 9, characterized in that, The path discretization module, inverse solution module, configuration screening module, marker configuration selection module, and path splicing module are configured to perform the steps of the CT imaging robot scanning path collaborative planning method as described in any one of claims 2 to 8.

Images