A robotic multi-object path planning system and method for an individual's head

Abstract

The application discloses a robot multi-target path planning system for an individual head, the system drives an end tool to move along a planned path through a mechanical arm, and the system comprises a head model receiving module, which receives a three-dimensional model of the individual head; a visual positioning module, which performs visual positioning to obtain regional pose data of the individual head in a base coordinate system of the mechanical arm; a head model registration module, which registers the individual head with the three-dimensional model to obtain target point pose data; and a head multi-target path planning module, which is used for acquiring a plurality of path curves in the three-dimensional model according to the target point pose data and taking a path curve corresponding to the shortest curve length as a planned path between two target points. Through the technical scheme provided by the application, the moving path of the end tool controlled by the robot is shorter, and the consistency with the surface profile of the individual head is better; in the process of advancing between the two target points, the posture of the end tool is adjusted in real time, and the safety of the end tool in the process of switching between the target points can be well ensured.

Description

Technical Field

[0001] This invention relates to the field of robot path planning technology, and in particular, to a multi-objective path planning system and method for a robot targeting an individual head. Background Technology

[0002] Using robotic systems to assist in localization and tracking offers significant advantages in accuracy, repeatability, and ease of use for operators. However, despite these advantages, current head path planning techniques face challenges. Individual differences in head shape and structure make it difficult to switch between multiple target points after path planning. The end effector cannot follow the same path for different individuals, hindering this process. Furthermore, improper path planning can lead to collisions with the individual's head, causing safety issues. Current research often uses spherical or ellipsoidal models to enclose the individual's head, causing the planned path to lie outside the envelope. Since the subject's head shape differs from the envelope model, this results in longer planned paths and makes end effector pose adjustment difficult. Summary of the Invention

[0003] The purpose of this invention is to provide a multi-objective path planning system and method for a robot head, in order to solve the problem of rationally planning the travel path of the end effector controlled by the robot.

[0004] To achieve the above objectives, this invention provides a multi-target path planning system for a robot head. The system uses a robotic arm to move an end effector along a planned path. The system includes: a head model receiving module for receiving a 3D model of the individual head contour; a visual positioning module for visually positioning the individual head to obtain regional pose data of the individual head in the robotic arm's base coordinate system; a head model registration module for registering the individual head with the 3D model based on the regional pose data to obtain target point pose data of multiple pre-set target points on the individual head in the robotic arm's base coordinate system; and a head multi-target path planning module for obtaining multiple path curves corresponding to two target points in the 3D model based on their target point pose data, calculating the curve lengths of the multiple path curves, and using the path curve corresponding to the shortest curve length as the planned path between the two target points.

[0005] Preferably, the head multi-target path planning module includes: a curve acquisition unit, used to acquire multiple path curves corresponding to two target points in the three-dimensional model based on the target pose data of two target points among the multiple target points; a length calculation unit, used to calculate the curve length of the multiple path curves; and a planning path selection unit, used to select the path curve with the shortest curve length among the multiple path curves as the planning path between the two target points.

[0006] Preferably, the curve acquisition unit is further configured to: determine multiple model curves on the three-dimensional model by multiple planes passing through the line connecting the two target points; and for each of the multiple model curves, extend and offset the model curve outward by a fixed distance to obtain the multiple path curves.

[0007] Preferably, the length calculation unit is further configured to: perform segmented interpolation on each of the multiple path curves to obtain multiple interpolation points on each path curve; calculate the pose data of the multiple interpolation points and the pose data of the start and end points corresponding to the two target points for each path curve based on the target pose data of the two target points; and calculate the curve length of the multiple path curves based on the pose data of the multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points.

[0008] Preferably, obtaining multiple interpolation points and their pose data on each path curve includes: taking multiple equally spaced intermediate points on the line connecting the two target points; for each of the multiple model curves, projecting the multiple intermediate points onto the multiple model curves through a plane perpendicular to the line connecting the two target points and passing through each intermediate point, thereby obtaining multiple projection points on each model curve; calculating the pose data of the multiple projection points on each model curve based on the target pose data of the two target points; and for each of the multiple model curves, taking the projection points along a plane perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection plane... The shadow lines map the multiple projection points onto the multiple path curves, obtaining multiple interpolation points on each path curve; for each of the multiple model curves, lines perpendicular to the tangent planes at the two target points on the model curve and passing through the corresponding target points map the two target points onto the multiple path curves, obtaining the start and end points corresponding to the two target points on each path curve; and based on the pose data of the multiple projection points on each model curve, the pose data of the two target points, and the fixed distance, the pose data of the multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points are calculated.

[0009] Accordingly, the present invention also provides a multi-target path planning method for a robot head, wherein a robotic arm drives an end effector held by the robotic arm to move along a planned path above the individual head. The method includes: receiving a three-dimensional model of the individual head; performing visual localization on the individual head to obtain regional pose data of the individual head in the robotic arm base coordinate system; registering the individual head with the three-dimensional model based on the regional pose data to obtain target pose data of multiple target points pre-set on the individual head in the robotic arm base coordinate system; and obtaining multiple path curves corresponding to two target points in the three-dimensional model based on the target pose data of the multiple target points, calculating the curve lengths of the multiple path curves, and using the path curve corresponding to the shortest curve length as the planned path between the two target points.

[0010] Preferably, obtaining multiple path curves in the 3D model based on the target pose data of two target points from the plurality of target points, calculating the curve lengths of the multiple path curves, and using the path curve with the shortest curve length as the planned path between the two target points includes: obtaining multiple path curves in the 3D model based on the target pose data of two target points from the plurality of target points; calculating the curve lengths of the multiple path curves; and selecting the path curve with the shortest curve length from the multiple path curves as the planned path between the two target points.

[0011] Preferably, obtaining multiple path curves in the 3D model based on the target pose data of two of the multiple target points includes: determining multiple model curves by connecting multiple planes through the line connecting the two target points with the 3D model; and for each of the multiple model curves, expanding and offsetting the model curve outward by a fixed distance to obtain the multiple path curves.

[0012] Preferably, calculating the curve length of the multiple path curves includes: performing piecewise interpolation on each of the multiple path curves to obtain multiple interpolation points on each path curve; calculating the pose data of the multiple interpolation points and the pose data of the start and end points corresponding to the two target points for each path curve based on the target pose data of the two target points; and calculating the curve length of the multiple path curves based on the pose data of the multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points.

[0013] Preferably, obtaining multiple interpolation points and their pose data on each path curve includes: taking multiple equally spaced intermediate points on the line connecting the two target points; for each of the multiple model curves, projecting the multiple intermediate points onto the multiple model curves through a plane perpendicular to the line connecting the two target points and passing through each intermediate point, thereby obtaining multiple projection points on each model curve; calculating the pose data of the multiple projection points on each model curve based on the target pose data of the two target points; and for each of the multiple model curves, taking the projection points along a plane perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection plane... The shadow lines map the multiple projection points onto the multiple path curves, obtaining multiple interpolation points on each path curve; for each of the multiple model curves, lines perpendicular to the tangent planes at the two target points on the model curve and passing through the corresponding target points map the two target points onto the multiple path curves, obtaining the start and end points corresponding to the two target points on each path curve; and based on the pose data of the multiple projection points on each model curve, the pose data of the two target points, and the fixed distance, the pose data of the multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points are calculated.

[0014] This invention registers the region pose data obtained through visual localization of the area to be multi-target path planning with the region's 3D model to obtain target pose data for multiple target points within the region. Then, based on the target pose data, multiple path curves corresponding to these target points are acquired in the 3D model, and the travel path between two target points is planned according to the path curve lengths. Through the technical solution provided by this invention, the robot-controlled end-effector moves along a shorter path with better consistency with the individual's head surface contour. The end-effector's posture is adjusted in real-time during its movement between two target points, effectively ensuring the safety of the end-effector during target point switching. Attached Figure Description

[0015] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0016] Figure 1 This is a block diagram of the multi-objective path planning system for an individual head robot provided by the present invention.

[0017] Figure 2 This is a block diagram of another robot multi-objective path planning system for an individual head provided by the present invention.

[0018] Figure 3This is a schematic diagram of the projection points, interpolation points, and related three-dimensional coordinates provided by the present invention.

[0019] Figure 4 This is a flowchart of the multi-objective path planning method for a robot targeting an individual head provided by the present invention. Detailed Implementation

[0020] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0021] Figure 1 This is a block diagram of the multi-objective path planning system for a robot targeting an individual head provided by the present invention, as shown below. Figure 1 As shown, the system includes a head model receiving module 10, a visual positioning module 20, a head model registration module 30, and a head multi-objective path planning module 40. The robot multi-objective path planning system for an individual head provided by this invention uses a robotic arm to move an end effector held by the robotic arm along a planned path above the individual's head.

[0022] The head model receiving module 10 is used to receive the three-dimensional model of an individual's head. When path planning is required for a region, the three-dimensional model of the individual's head must first be established.

[0023] The visual positioning module 20 is used to perform visual positioning of the individual's head to obtain the regional pose data of the individual's head in the robot arm's base coordinate system. Since the individual's head may be displaced during the path planning process, it is necessary to use the visual positioning module 20 to position the individual's head and obtain the pose data (i.e., regional pose data) of the individual's head based on the robot arm's base coordinate system.

[0024] The head model registration module 30 is used to register the individual's head with the 3D model based on the region pose data, so as to obtain the target pose data of multiple target points pre-set on the individual's head in the robot arm's base coordinate system. Multiple target points are pre-set on the individual's head. After path planning is completed, the robot arm will drive its gripping end effector to sequentially pass through the pre-set target points. After the vision positioning module 20 obtains the region pose data in the robot arm's base coordinate system, the head model registration module 30 registers the individual's head with the 3D model of the individual's head received by the head model receiving module 10, thereby obtaining the target pose data of multiple target points on the individual's head. The registration process performed by the head model registration module 30 and the process of obtaining the target pose data of the pre-set multiple target points can be implemented using conventional techniques and will not be elaborated here.

[0025] The head multi-target path planning module 40 is used to obtain multiple path curves in the 3D model based on the target pose data of two targets out of multiple target points, calculate the curve lengths of the multiple path curves, and use the path curve corresponding to the shortest curve length as the planned path between the two target points. In the 3D model, multiple path curves passing through two target points can be determined, and the path curve corresponding to the shortest path curve among these multiple path curves is used as the planned path between the two target points.

[0026] Figure 2 This is a block diagram of another multi-objective path planning system for a robot targeting an individual head, provided by the present invention, such as... Figure 2 As shown, the head multi-objective path planning module 40 includes a curve acquisition unit 41, a length calculation unit 42, and a planned path selection unit 43.

[0027] The curve acquisition unit 41 is used to acquire multiple path curves corresponding to two target points in the 3D model based on the target pose data of two target points out of multiple target points. After the multi-target path planning is completed, the planned path generally passes through multiple pre-set target points in a pre-set order.

[0028] More specifically, the curve acquisition unit 41 is also used to: determine multiple model curves on the three-dimensional model by multiple planes that pass through the line connecting the two target points; and for each of the multiple model curves, extend the model curve outward by a fixed distance to obtain multiple path curves.

[0029] The line connecting two target points can form an infinite number of planes. Multiple planes can be obtained by rotating around this line with an arbitrary step size. In this embodiment, a 1° step size rotation yields 360 planes. The curves formed by the intersection of these 360 ​​planes with the 3D contour of the 3D model are the determined model curves. In this embodiment, 360 model curves are determined. To prevent the end effector from colliding with the individual's head, the determined model curves (360 model curves in this embodiment) are offset outward by a fixed distance (e.g., 1cm) to obtain corresponding path curves. In this embodiment, 360 path curves are obtained corresponding to the 360 ​​model curves. This outward offset technique is well-known in the art and will not be elaborated upon here. The purpose of the 1cm outward offset is to ensure that the end effector does not contact the surface of the individual's head and maintains a fixed distance from the individual's head surface during its movement.

[0030] The length calculation unit 42 is used to calculate the curve length of multiple path curves. Those skilled in the art can calculate the curve length using existing techniques, or they can calculate it using the technical solution provided in this invention.

[0031] The path selection unit 43 is used to select the path curve with the shortest curve length from multiple path curves as the planned path between two target points. Generally, a complete planned path will pass through multiple target points, and the order of the target points is preset. For example, if a complete path has three target points, namely the first target point, the second target point, and the third target point, then the preset target point order is from the first target point to the second target point, and then from the second target point to the third target point. Here, the first target point and the second target point are adjacent targets, and the second target point and the third target point are also adjacent targets.

[0032] For each pair of adjacent target points, path planning is required. Once the paths between each pair of adjacent target points are planned, a complete planned path will be formed. Since the path planning method between any two adjacent target points is the same, this invention only focuses on describing the path planning process between two target points.

[0033] The length calculation unit 42 is also used to: perform segmented interpolation on each of the multiple path curves to obtain multiple interpolation points on each path curve; calculate the pose data of the multiple interpolation points and the pose data of the start and end points corresponding to the two target points for each path curve based on the target pose data of the two target points; and calculate the curve length of the multiple path curves based on the pose data of the multiple interpolation points and the pose data of the start and end points corresponding to the two target points.

[0034] To calculate the length of each path curve, piecewise interpolation is required for each path curve. This involves interpolating between two target points on the path curve to obtain multiple interpolation points. The selection of interpolation points can be arbitrary or set according to certain rules, such as setting an interpolation point at fixed intervals. Given the target pose data of two target points and multiple path curves, the pose data of multiple interpolation points on each path curve can be calculated according to the rules of piecewise interpolation. It should be understood that the pose data includes three-dimensional coordinates. Here, the distance between each two adjacent interpolation points between two target points can be calculated based on the three-dimensional coordinates in the pose data of multiple interpolation points. Summing these distances gives the curve length between the two target points. The curve length of each determined path curve is calculated, thereby obtaining the curve length of multiple path curves.

[0035] This invention provides a technical solution for performing segmented interpolation on a path curve and obtaining pose data of interpolation points. Specifically, obtaining multiple interpolation points and pose data of multiple interpolation points on each path curve includes: taking multiple equally spaced intermediate points on the line connecting two target points; for each of the multiple model curves, projecting the multiple intermediate points onto the multiple model curves using a plane perpendicular to the line connecting the two target points and passing through each intermediate point, thereby obtaining multiple projection points on each model curve; calculating the pose data of the multiple projection points on each model curve based on the target pose data of the two target points; and for each of the multiple model curves... For each model curve, a line perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection point is used to map multiple projection points onto multiple path curves, resulting in multiple interpolation points on each path curve. For each model curve, a line perpendicular to the tangent plane at two target points on the model curve and passing through the corresponding target points is used to map two target points onto multiple path curves, resulting in the start and end points corresponding to the two target points on each path curve. Based on the pose data of multiple projection points on each model curve, the pose data of the two target points, and a fixed distance, the pose data of multiple interpolation points on each path curve, as well as the pose data of the start and end points corresponding to the two target points, are calculated.

[0036] This invention uses the switching between two target points out of multiple target points as an example for illustration. It should be understood that the robot-controlled end effector switches between multiple target points in a pre-set order. The path planned in this embodiment is the switching of the end effector from the first target point to the second target point out of two target points. The first target point is denoted by Ts, and its pose data is Ts = [Xs, Ys, Zs, αs, βs, γs]. The second target point is denoted by Te, and its pose data is Te = [Xe, Ye, Ze, αe, βe, γe]. N-1 points are taken at equal intervals on the line connecting the first target point Ts and the second target point Te, where N is a positive integer. These N-1 points are the intermediate points. The pose data of the first target point Ts and the second target point Te are known, and the pose data of the N-1 intermediate points can be calculated. Here, the pose data of the first and second target points are both data in the robot arm's base coordinate system.

[0037] Then, for each model curve, multiple intermediate points are projected onto multiple model curves using a plane perpendicular to the line connecting the first target point Ts and the second target point Te, passing through each of the N-1 intermediate points. This results in multiple projection points on each model curve. Specifically, for one model curve, N-1 points are projected onto that curve using a plane perpendicular to the line connecting the first target point Ts and the second target point Te, passing through each of the N-1 intermediate points. Therefore, N-1 projection points are obtained for each model curve.

[0038] Given the pose data of N-1 intermediate points, and since each model curve is also known, the pose data of N-1 projection points can be calculated. These N-1 projection points can be, for example, T1, T2, ..., T... i , ......, T N-1 Indicates that the projection point T i The pose data can be represented as T i =[Xi, Yi, Zi, αi, βi, γi], where, T i The y-axis in the figure represents the corresponding model curve in T. i The tangent to the point, with the y-axis direction being the direction from the first target point through the corresponding model curve to the second target point, T i The z-axis in the equation is perpendicular to the line passing through T. i The line of the tangent plane at the point has its z-axis pointing inwards (this inwards corresponds to the outward expansion in the path curve formed by expanding the model curve outwards). For details, please refer to [reference needed]. Figure 3 , Figure 3 This is a schematic diagram of the projection points, interpolation points, and related three-dimensional coordinates provided by this invention. It should be understood that the directions of the x-axis, y-axis, and z-axis of each projection point are different, but the setting rules are the same.

[0039] After obtaining N-1 projection points and their pose data, the N-1 projection points are mapped onto the corresponding path curves along lines perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection point. Simply put, this means moving these N-1 projection points a fixed distance h (e.g., h = 1 cm) along the -z axis direction corresponding to each projection point. Then, for each path curve, N-1 interpolation points can be obtained, denoted by K1, K2, ..., K... i , ......, K N-1 This means that, given the pose data of N-1 projection points, the pose data of N-1 interpolation points can be obtained, where K is the number of interpolation points. iThe pose data can be represented as Ki = [Xi, Yi, Zi-h, αi, βi, γi]. The pose data of the first target point Ts and the second target point Te mapped to the corresponding start and end points on the path curve are calculated in the same way as the pose data of the interpolation points. That is, the start point corresponding to the first target point Ts is moved a fixed distance h along the -z axis direction corresponding to the first target point Ts, and the end point corresponding to the second target point Te is moved a fixed distance h along the -z axis direction corresponding to the second target point Te.

[0040] For a path curve, the sum of the straight-line distances between any two adjacent points of the N-1 interpolation points between the starting point corresponding to the first target point Ts and the ending point corresponding to the second target point Te, and the sum of the straight-line distances between any two adjacent points of the N+1 interpolation points between the starting point corresponding to the first target point Ts and the ending point corresponding to the second target point Te, can be represented as d1, ..., d2. i , ......, d N It can be understood that there are N line segments in total, and the calculation formula is as follows:

[0041]

[0042]

[0043]

[0044] In formula (1), d1 represents the distance from the starting point corresponding to the first target point Ts to the interpolation point T1. i Indicates the interpolation point T i-1 to interpolation point T i The distance, d N Indicates the interpolation point T N-1 The distance to the endpoint corresponding to the second target point Te.

[0045] The formula for calculating curve length is as follows:

[0046]

[0047] In formula (4), D represents the length of a path curve.

[0048] For each path curve, a curve length D is calculated, resulting in multiple curve lengths for multiple path curves. The path selection unit 43 compares the curves to find the shortest curve length and uses the path curve corresponding to the shortest curve length as the planned path between the first target point Ts and the second target point Te.

[0049] Figure 4This is a flowchart of a multi-objective path planning method for a robot targeting an individual's head, provided by the present invention. The method involves a robotic arm driving an end effector held by the arm to move along a planned path above the individual's head. Figure 4 As shown, the method includes:

[0050] Step S401: Receive the three-dimensional model of the individual's head;

[0051] Step S402: Visually locate the individual head to obtain the regional pose data of the individual head in the robot arm base coordinate system;

[0052] Step S403: Register the individual head with the 3D model based on the regional pose data to obtain the target pose data of multiple target points pre-set on the individual head in the robot arm base coordinate system.

[0053] Step S404: Based on the target pose data of two target points out of multiple target points, obtain the corresponding multiple path curves in the 3D model, calculate the curve length of the multiple path curves, and take the path curve corresponding to the shortest curve length as the planned path between the two target points.

[0054] The process of obtaining multiple path curves in a 3D model based on the target pose data of two target points from a plurality of target points, calculating the curve lengths of the multiple path curves, and selecting the path curve with the shortest curve length as the planned path between the two target points includes: obtaining multiple path curves in a 3D model based on the target pose data of two target points from a plurality of target points; calculating the curve lengths of the multiple path curves; and selecting the path curve with the shortest curve length from the multiple path curves as the planned path between the two target points.

[0055] The process of obtaining multiple path curves in the 3D model based on the target pose data of two target points out of multiple target points includes: determining multiple model curves by connecting multiple planes through the line between the two target points and the 3D model; and for each model curve, expanding and offsetting the model curve outward by a fixed distance to obtain multiple path curves.

[0056] The calculation of the curve length of multiple path curves includes: performing piecewise interpolation on each of the multiple path curves to obtain multiple interpolation points on each path curve; calculating the pose data of the corresponding multiple interpolation points and the pose data of the start and end points of the two target points for each path curve based on the target pose data of the two target points; and calculating the curve length of the multiple path curves based on the pose data of the multiple interpolation points and the pose data of the start and end points of the two target points.

[0057] The process of obtaining multiple interpolation points and their pose data on each path curve includes: taking multiple equally spaced intermediate points on the line connecting two target points; for each of the multiple model curves, projecting the multiple intermediate points onto the multiple model curves using a plane perpendicular to the line connecting the two target points and passing through each intermediate point, thus obtaining multiple projection points on each model curve; calculating the pose data of the multiple projection points on each model curve based on the target pose data of the two target points; and for each of the multiple model curves, taking multiple projection points along a plane perpendicular to the tangent plane at each projection point of the model curve and passing through each intermediate point. Multiple projection points are mapped onto multiple path curves by lines passing through the corresponding projection points, resulting in multiple interpolation points on each path curve. For each of the multiple model curves, two target points are mapped onto multiple path curves by lines perpendicular to the tangent planes at the two target points on the model curve and passing through the corresponding target points, resulting in the start and end points corresponding to the two target points on each path curve. Based on the pose data of multiple projection points on each model curve, the pose data of the two target points, and a fixed distance, the pose data of multiple interpolation points on each path curve, as well as the pose data of the start and end points corresponding to the two target points, are calculated.

[0058] It should be noted that the specific details and benefits of the robot multi-objective path planning method for individual heads provided by this invention are similar to those of the robot multi-objective path planning system for individual heads provided by this invention, and will not be repeated here.

[0059] The optional embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the scope of the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all fall within the protection scope of the embodiments of the present invention.

[0060] The technical solution provided by this invention can shorten the path of the end-effector from one target point to another. During the process from the starting target point to the target target point, the distance between the end-effector and the surface of the individual's head remains unchanged and the posture is adjusted in real time. When it reaches the target target point, its posture does not need to be adjusted to meet the requirements. It has better adaptability to the shape of the individual's head and also ensures the safety of the end-effector during the process of moving to multiple target points on the individual's head.

[0061] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not describe the various possible combinations separately.

[0062] Furthermore, various different implementations of the present invention can be combined arbitrarily, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed in the present invention.

Claims

1. A multi-objective path planning system for a robot targeting an individual's head, characterized in that, The system uses a robotic arm to move an end effector held by the robotic arm along a planned path above the individual's head. The system includes: A head model receiving module is used to receive a three-dimensional model of the individual's head; A visual positioning module is used to perform visual positioning of the individual's head to obtain the regional pose data of the individual's head in the robot arm's base coordinate system. A head model registration module is used to register the individual's head with the 3D model based on the region pose data, so as to obtain target point pose data of multiple pre-set target points on the individual's head in the robot arm's base coordinate system; and The head-based multi-target path planning module is used to obtain multiple path curves in the 3D model based on the target pose data of two targets among the multiple target points, calculate the curve lengths of the multiple path curves, and use the path curve corresponding to the shortest curve length as the planned path between the two target points. The head-based multi-objective path planning module includes: The curve acquisition unit is used to acquire multiple path curves in the three-dimensional model based on the target pose data of two target points among the multiple target points. A length calculation unit is used to calculate the curve lengths of the multiple path curves; and The path selection unit is used to select the path curve with the shortest curve length from the multiple path curves as the planned path between the two target points. Furthermore, the curve acquisition unit is also used for: Multiple planes connected by the two target points respectively define multiple model curves on the three-dimensional model; and For each of the multiple model curves, the model curve is shifted outward by a fixed distance to obtain the multiple path curves.

2. The multi-objective path planning system for a robot targeting an individual head according to claim 1, characterized in that, The length calculation unit is also used for: For each of the multiple path curves, segmented interpolation is performed to obtain multiple interpolation points on each path curve. Based on the target pose data of the two target points, calculate the pose data of multiple interpolation points for each path curve, as well as the pose data of the start and end points corresponding to the two target points. as well as The curve lengths of the multiple path curves are calculated based on the pose data of multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points.

3. The multi-objective path planning system for a robot targeting an individual head according to claim 2, characterized in that, On each path curve, multiple interpolation points and their pose data are obtained, including: Take multiple equally spaced intermediate points on the line connecting the two target points; For each of the multiple model curves, the multiple intermediate points are projected onto the multiple model curves by means of a plane that is perpendicular to the line connecting the two target points and passes through each intermediate point, thereby obtaining multiple projection points on each model curve. Based on the target pose data of the two target points, the pose data of multiple projection points on each model curve are calculated. For each of the multiple model curves, the multiple projection points are mapped onto the multiple path curves along a line perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection point, thereby obtaining multiple interpolation points on each path curve. For each of the multiple model curves, the two target points are mapped onto the multiple path curves along lines perpendicular to the tangent planes at the two target points on the model curve and passing through the corresponding target points. The starting point and ending point corresponding to the two target points are obtained on each path curve. Based on the pose data of multiple projection points on each model curve, the pose data of the two target points, and the fixed distance, the pose data of multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points are calculated.

4. A multi-objective path planning method for a robot targeting an individual head, characterized in that, Using the multi-objective path planning system for an individual's head as described in claim 1, a robotic arm drives an end effector held by the robotic arm to move along a planned path above the individual's head. The method includes: Receive the three-dimensional model of the individual's head; Visual localization of the individual's head is performed to obtain the regional pose data of the individual's head in the robot arm's base coordinate system; The individual's head is registered with the 3D model based on the region pose data to obtain target point pose data for multiple pre-set target points on the individual's head in the robotic arm's base coordinate system; and Based on the target pose data of two of the multiple target points, obtain the corresponding multiple path curves in the 3D model, calculate the curve length of the multiple path curves, and take the path curve corresponding to the shortest curve length as the planned path between the two target points.

5. The multi-objective path planning method for a robot targeting an individual head according to claim 4, characterized in that, Based on the target pose data of two target points from the plurality of target points, obtain multiple path curves corresponding to the target points in the 3D model, calculate the curve lengths of the multiple path curves, and use the path curve corresponding to the shortest curve length as the planned path between the two target points, including: Based on the target pose data of two target points out of the plurality of target points, obtain the corresponding multiple path curves in the three-dimensional model; Calculate the curve lengths of the multiple path curves; and The path curve with the shortest curve length among the multiple path curves is selected as the planned path between the two target points.

6. The multi-objective path planning method for a robot targeting an individual head according to claim 5, characterized in that, Based on the target pose data of two target points out of the plurality of target points, the corresponding multiple path curves in the 3D model are obtained, including: Multiple planes connected by the line between the two target points respectively define multiple model curves with the three-dimensional model; and For each of the multiple model curves, the model curve is shifted outward by a fixed distance to obtain the multiple path curves.

7. The multi-objective path planning method for a robot targeting an individual head according to claim 6, characterized in that, Calculating the curve lengths of the multiple path curves includes: For each of the multiple path curves, segmented interpolation is performed to obtain multiple interpolation points on each path curve. Based on the target pose data of the two target points, calculate the pose data of multiple interpolation points for each path curve, as well as the pose data of the start and end points corresponding to the two target points; and The curve lengths of the multiple path curves are calculated based on the pose data of multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points.

8. The multi-objective path planning method for a robot targeting an individual head according to claim 7, characterized in that, On each path curve, multiple interpolation points and their pose data are obtained, including: Take multiple equally spaced intermediate points on the line connecting the two target points; For each of the multiple model curves, the multiple intermediate points are projected onto the multiple model curves by means of a plane that is perpendicular to the line connecting the two target points and passes through each intermediate point, thereby obtaining multiple projection points on each model curve. Based on the target pose data of the two target points, the pose data of multiple projection points on each model curve are calculated. For each of the multiple model curves, the multiple projection points are mapped onto the multiple path curves along a line perpendicular to the tangent plane at each projection point of the model curve and passing through the corresponding projection point, thereby obtaining multiple interpolation points on each path curve. For each of the multiple model curves, the two target points are mapped onto the multiple path curves along lines perpendicular to the tangent planes at the two target points on the model curve and passing through the corresponding target points. The starting point and ending point corresponding to the two target points are obtained on each path curve. Based on the pose data of multiple projection points on each model curve, the pose data of the two target points, and the fixed distance, the pose data of multiple interpolation points on each path curve and the pose data of the start and end points corresponding to the two target points are calculated.

Images