Coordinate system construction method, prism scanner multi-target detection scanning method and system

Abstract

The present application relates to a kind of coordinate system construction method, prism scanner multi-target detection scanning method and system.Compared with prior art, the present application constructs multiple prism coordinate system, the motion angle of each prism of cascade prism is assigned to the coordinate system axis with same dimension, a coordinate point represents the prism state at a certain time, the distance between coordinate points represents the prism motion in a period of time;Both realize the complete description of prism motion state, and reduce the definition and use difficulty of coordinate system, convert the space target point planning problem into directly control the prism motion path planning of laser beam, fast, efficient, the path obtained by planning is to realize the shortest prism running distance or the fastest speed as scanning target, reduce the overall energy consumption of scanner, improve the smoothness of prism running, and the same direction principle can also avoid motor eccentric wear, provide theoretical basis and application foundation for large-scale, long-distance, disordered multi-target scanning tracking.

Description

Technical Field

[0001] This invention relates to the field of prism scanner technology, and in particular to a coordinate system construction method, a multi-target detection scanning method and system for prism scanners. Background Technology

[0002] Active photoelectric recognition scanning systems are based on optical axis adjustment technology. This involves continuously adjusting the direction of the visual axis to ensure that the visual sensor's pointing endpoint is the target being identified, or to keep the target of interest always in the center of the field of view. Researchers acquire multi-view image information by changing the pose of the visual sensor or altering the imaging visual axis angle. Research revealed that schemes for changing the imaging axis angle of a camera can be divided into three categories: multi-axis gimbals (Yang Liang, Zhou Yong, Liu Xinliu. Moving target tracking control based on PTZ camera [J]. Ordnance Automation, 2014, (3): 65-68.), reflective optical systems (Shanghai Institute of Technical Physics, Chinese Academy of Sciences. Fast reflective mirror scanning tracking system and method for aerospace imaging: CN201410020863.5 [P]. 2014-06-25.), and refractive optical systems (Cui XY, Zhao Y., Lim KB, et al. Perspective projection model for prism-based stereovision. Optics Express, 2015, 23: 27542–27557.63.). Among them, the refractive optical variable-axis imaging scheme has a fixed relative pose of the visual sensor and changes the imaging axis by the swaying or rotation of the front wedge optical prism. It has the characteristics of fast response, high precision, compact structure, and good robustness. Prism-based lidar has been applied in fields such as laser communication, radar, and infrared countermeasures.

[0003] When a prism scanner is in operation, the rotation angle and rotation speed of each prism determine the pointing and motion state of the adjustable optical axis. Therefore, characterizing the state of each prism in the prism scanner is a necessary condition for active visual recognition and photoelectric tracking. However, common Cartesian coordinate systems, planar polar coordinate systems, cylindrical coordinate systems, and spherical coordinate systems cannot intuitively and simultaneously express the rotation angle position of multiple prisms at time t, or the position change from time t to time t+1. Furthermore, the spatial optical axis pointing vector of the prism scanner and the rotation angle value of each prism have a high degree of nonlinearity and strong coupling, which means that the spatial pointing scanning plan in the traditional Cartesian coordinate system is not the optimal solution for prism motion, and the two are not linearly related.

[0004] Due to the limitations of the aforementioned coordinate system, prism scanners cannot fully represent the motion path of the prism assembly during multi-target detection scanning, making it difficult to achieve optimal scanning path planning. There is an urgent need for a novel coordinate system to describe the motion of the cascaded prism assembly, characterize its relationship with the line-of-sight adjustment, and apply it to optimal scanning path planning during multi-target detection scanning. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a coordinate system construction method, a prism scanner multi-target detection and scanning method and system, which assigns prism motion values ​​to Cartesian coordinate axes to characterize the prism's motion state. Within this coordinate system, the prism's operating path can be directly planned, achieving fast, low-energy-consumption unordered multi-target scanning.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A coordinate system construction method, applied to a cascaded prism assembly, includes:

[0008] Establish an n-dimensional coordinate system, where n equals the number of prisms in the cascaded prism group. Define points in the coordinate system as follows: each coordinate axis corresponds to the angle of independent motion of different prisms, and a point on a coordinate axis in the coordinate system represents the state of a prism at a certain moment.

[0009] Defining the distance between points in a coordinate system: The distance between two points in a coordinate system is expressed as:

[0010] In the shortest distance mode:

[0011]

[0012]

[0013] Where, θ ia θ represents the projection of point a onto coordinate axis i. ib This represents the projection of point b onto coordinate axis i. This represents the distance between point a and point b across all coordinate axes, mathematically defined as the Manhattan distance, where points a and b represent time t. a And the next moment t b The prism state, d ml This represents the total distance traveled by the prism when there are l target points in the shortest distance mode.

[0014] Because the prism rotates simultaneously, in the fastest distance mode:

[0015]

[0016]

[0017] D represents the maximum distance traveled between points a and b, mathematically defined as the Chebyshev distance. cl This represents the total distance traveled by the prism at the fastest scan speed when there are l target points in the fastest distance mode.

[0018] Define the path difference between points in the coordinate system:

[0019]

[0020]

[0021] in, D represents the maximum path difference between points a and b; d This represents the maximum total path difference of the prism when there are l target points.

[0022] Furthermore, for ease of calculation, the maximum difference ratio is used:

[0023]

[0024] I d This represents the maximum difference ratio between l target points.

[0025] A multi-target detection scanning method using a prism scanner, the prism scanner comprising a lidar transmitter, a visually guided rangefinder camera, a cascaded prism group, and a support and control system, wherein the visually guided rangefinder camera is mounted off-axis of the lidar transmitter, and the cascaded prism group is coaxially and parallel in planar arrangement in front of the lidar transmitter; the scanning method includes the following steps:

[0026] S1. Calibrate the camera intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the pixel coordinate system; use the visually guided rangefinder camera to take pictures of multiple targets at a distance, identify and mark the targets in the acquired images, and obtain a target point map in the pixel coordinate system.

[0027] S2. Use the coordinate system construction method described above to establish a multi-prism coordinate system, where the motion angle of the prism is a function of time t;

[0028] S3. Establish the inverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism to realize the transformation from the pixel coordinate system to the multi-prism coordinate system;

[0029] S4. In the multiprism coordinate system, the motion planning of the prism scanner is abstracted into a multi-point traveling salesman problem, and the motion path of the multiprism is planned.

[0030] S5. Based on the planned multi-prism motion path, control the prism scanner to scan each target point sequentially.

[0031] Furthermore, it also includes:

[0032] S6. Based on the planned multi-prism motion path, the outgoing laser direction is calculated using the iterative vector method, and a predicted target point map is formed in the world coordinate system. The predicted target point map is compared with the target point map obtained from the actual scanning direction in step S5, and the prism scanner motion is corrected.

[0033] Furthermore, in step S1, the camera intrinsic parameters are calibrated using the Zhang Zhengyou method.

[0034] Furthermore, in step S4, the optimization objective is to find the shortest path, the fastest path, or a weighted combination of the two.

[0035] Furthermore, in step S4, the cascaded prism group generates 2n motion modes Mode, where n is the number of prisms.

[0036] Furthermore, in step S4, for the obtained multi-prism motion path, a prism motion path feature map is drawn according to the same-direction operation rule, and the shortest distance, the fastest distance, and the maximum path difference ratio are used as the evaluation criteria for the path.

[0037] A prism scanner multi-target detection scanning system, wherein the prism scanner includes a lidar transmitter, a visually guided rangefinder camera, a cascaded prism group, and a support control system. The visually guided rangefinder camera is mounted on the side axis of the lidar transmitter, and the cascaded prism group is coaxial and parallel in plane, positioned directly in front of the lidar transmitter. The scanning system includes:

[0038] The pixel coordinate system module is used to calibrate the camera's intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the pixel coordinate system. The visually guided rangefinder camera is used to photograph multiple targets at a distance, and the targets in the acquired images are identified and marked to obtain a target point map in the pixel coordinate system.

[0039] The multiprism coordinate system module is used to establish a multiprism coordinate system using the coordinate system construction method described above, where the motion angle of the prism is a function of time t.

[0040] The coordinate transformation module is used to establish the inverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism, and realize the transformation from the pixel coordinate system to the multi-prism coordinate system;

[0041] The planning module is used to abstract the motion planning of the prism scanner into a multi-point traveling salesman problem in the multi-prism coordinate system and plan the motion path of the multi-prism.

[0042] The scanning control module is used to control the prism scanner to sequentially scan each target point according to the planned multi-prism motion path.

[0043] Furthermore, it also includes:

[0044] The feedback correction module is used to calculate the direction of the emitted laser based on the planned multi-prism motion path using the iterative vector method, form a predicted target point map in the world coordinate system, compare the predicted target point map with the target point map obtained by the actual scanning direction in the scanning control module, and correct the prism scanner motion of the scanning control module.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] (1) Assign the motion angles of each cascaded prism to the axes of a coordinate system with the same dimension. A coordinate point represents the state of the prism at a certain moment, and the distance between coordinate points represents the motion of the prism over a period of time. This not only achieves a complete description of the motion state of the prism, but also follows the characteristics of the classic rectangular coordinate system. The motion of the prism can be defined by relevant mathematical distances, which greatly reduces the difficulty of defining and using the coordinate system.

[0047] (2) Based on the multi-prism coordinate system, the inverse solution of the prism is cleverly used to put the pointing vector of the spatial optical axis into the multi-prism coordinate system, and the spatial target point planning problem is transformed into the prism motion path planning that directly controls the laser beam. The spatial target point planning and the prism motion planning are decoupled, which is fast and efficient. The prism output pointing deceleration ratio is large and the pointing accuracy is high.

[0048] (3) The planned path aims to achieve the shortest prism running distance or the fastest speed, thereby reducing the overall energy consumption of the scanner, reducing the waiting time of different prisms, improving the smoothness of prism operation, and avoiding motor wear by adhering to the same direction principle. This provides a theoretical basis and application foundation for scanning and tracking large-scale, long-distance, and disordered multi-target targets. Attached Figure Description

[0049] Figure 1 This refers to the pointing planning process of a prism scanner for multiple targets in a spatially disordered environment.

[0050] Figure 2 This represents the correspondence between the cascaded prism group and the multi-prism coordinate system;

[0051] Figure 3 The transformation relationship between the camera pixel coordinate system and the multiprism coordinate system is shown in the example of a rotating dual-prism scanner; where (a) and (b) are the target point maps in the pixel coordinate system and the prism coordinate system, respectively.

[0052] Figure 4 Taking a rotating dual-prism scanner as an example, the path feature maps of two scanning targets and four motion modes in a multi-prism coordinate system are shown; where (a) is the shortest path planning and (b) is the fastest path planning.

[0053] Figure 5 To compare the path planning results using a multi-prism coordinate system and a traditional coordinate system, taking a rotating dual-prism scanner as an example;

[0054] Figure 6 This is a flowchart illustrating the specific application process of the rotating prism coordinate system described in this invention.

[0055] Figure labels: 1-Cascaded prism group; 2-LiDAR transmitter; 3-Multiple prism coordinate system; 4-Visual-guided ranging camera; 5-Camera pixel coordinate system; 6-Prism scanning field of view coordinate system (world coordinate system); 7-Scanning target. Detailed Implementation

[0056] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0057] Example 1:

[0058] A coordinate system construction method, applied to a cascaded prism assembly, includes:

[0059] (1) Establish an n-dimensional coordinate system, where n equals the number of prisms in the cascaded prism group. Define points in the coordinate system: each coordinate axis corresponds to the angle of independent motion of different prisms. A point on a coordinate axis in the coordinate system represents the state of a prism at a certain moment. Assign the motion angles of the n cascaded prisms to a coordinate system with the same dimension n. The orientation of the entire prism group can be decomposed into the motion of each prism. The motion angle of each prism is a function of time t. Using the n-dimensional coordinate system, the motion of the prism group over a period of time can be discretized into multiple points. Each point can represent the orientation state of the prism group at a certain moment. The coordinates of the point on each coordinate axis in the coordinate system represent the motion angles of each prism in the prism group.

[0060] (2) Define the distance between points in the coordinate system:

[0061] In the shortest distance mode:

[0062] The distance traveled by the prism assembly is the sum of the distances traveled by each individual prism. Therefore, the distance between two points is the distance traveled by the prism assembly from one state to another, expressed as the algebraic sum of the differences between the projections of the two points onto the coordinate axes, mathematically represented as the Manhattan distance. The distance between two points in the coordinate system is expressed as:

[0063]

[0064] Where, θ ia θ represents the projection of point a onto coordinate axis i.ib This represents the projection of point b onto coordinate axis i. This represents the distance between point a and point b, mathematically defined as the Manhattan distance, where points a and b represent time t. a And the next moment t b The prism state;

[0065] Therefore, for scanning l unordered target points in space, the total travel distance of the prism assembly during l state transitions is as follows:

[0066]

[0067] D ml This represents the total distance traveled by the prism when there are l unordered target points in the shortest distance mode.

[0068] In the fastest distance mode:

[0069] Since the movements of different prisms in the prism group are relatively independent, meaning that multiple prisms in the prism group can perform angular movements in parallel, the motion time of the prism group during the switching between the two states depends only on the maximum value of the difference between the projections of the two state points on the coordinate axis, mathematically expressed as the Chebyshev distance. Therefore, in the fastest distance mode, the total distance traveled by the prism scanning is as follows:

[0070]

[0071]

[0072] D represents the maximum distance traveled between points a and b, mathematically defined as the Chebyshev distance. cl This represents the total distance traveled by the prism when there are l unordered target points in the fastest distance mode.

[0073] (3) Define the path difference between points in the coordinate system:

[0074] Between two state points, the maximum distance the prism assembly travels on the coordinate axis is the Chebyshev distance, corresponding to a relatively shortest distance min. 1≤i≤n |θ ia -θ ib | This defines the maximum path difference of the prism assembly.

[0075]

[0076] in, This represents the difference in path between point a and point b;

[0077] Similarly, for l unordered target points, the maximum total path difference of the prism is defined as follows:

[0078]

[0079] Among them, D d This represents the maximum total path difference of a prism with l unordered target points.

[0080] Considering the maximum total difference D d The calculation method is rather complex; therefore, the maximum difference ratio is used to simplify the calculation.

[0081]

[0082] I d This represents the maximum difference ratio between l target points.

[0083] like Figure 1 As shown, the prism scanner includes a lidar transmitter, a vision-guided ranging camera, a cascaded prism assembly, and a support control system, and can be used for scanning path planning of disordered multiple targets.

[0084] A lidar transmitter emits a laser, which is refracted by a prism array and directed at a target in space. The scanner sequentially scans each target within the prism scanner's field of view coordinate system (world coordinate system).

[0085] The visually guided ranging camera is mounted on the off-axis of the lidar transmitter. Employing a high-frame-rate CCD camera, it captures multiple discrete, unordered targets in the world space and images them in the camera's pixel coordinate system. The camera's internal parameters, such as field of view, focal length, and resolution, as well as the external positional parameters (rotation and translation matrices) of the camera and its optical components, are adjusted synchronously according to the specific application scenario.

[0086] A cascaded prism assembly, coaxial and parallel in planar configuration, is positioned directly in front of the lidar transmitter. This assembly is used to adjust the laser direction. The cascaded prism assembly consists of multiple coaxial wedge-shaped prisms connected in series, and their movement includes independent rotation or independent yaw. The specific optical parameters and arrangement of the optical elements are matched and adjusted according to requirements such as the field of view.

[0087] The support control system includes a series of support, adjustment, transmission, drive, and control components. The drive components can employ direct drive of a torque motor, gear transmission, synchronous belt transmission, worm gear transmission, or other methods.

[0088] This application provides a multi-target detection scanning method using a prism scanner, involving three main coordinate systems: the world coordinate system (prism scanner field of view coordinate system), the camera pixel coordinate system, and the multi-prism coordinate system. The coordinate points in the multi-prism coordinate system represent the prism angle state at a given moment, and the distance between coordinate points represents the prism motion over a period of time. For unordered multi-target scanning problems, the scanning method includes the following steps:

[0089] S1. Constructing the camera imaging model: Use Zhang Zhengyou's method to calibrate the camera's intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the camera pixel coordinate system.

[0090] Creating a pixel target point map: A visually guided ranging camera is used to photograph multiple targets at a distance. Based on the images acquired by the camera, a visual recognition algorithm is used to identify and mark the feature parts, which are then abstracted and simplified into a target point map in a pixel coordinate system.

[0091] S2. Establish a multiprism coordinate system: Use the coordinate system construction method described above to establish a multiprism coordinate system, where the motion angle of the prism is a function of time t;

[0092] like Figure 2 As shown, the prism scanner consists of i coaxial prisms connected in series. The prisms move in two ways: rotation and yaw, with a motion angle of θ. A multi-prism coordinate system assigns the magnitude of the motion angle θ of the i cascaded prisms to rectangular coordinate axes with the same dimension n, where n = i (i ∈ Z+). Each coordinate axis represents the independent motion of the corresponding prism, with the motion angle being a function of time t. i (t).

[0093] S3. Construct a reverse calculation model for prism motion: Establish the reverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism, realize the transformation from the pixel coordinate system to the multiprism coordinate system, thereby realizing the coordinate transformation between the world coordinate system (prism scanner field of view coordinate system), the camera pixel coordinate system, and the multiprism coordinate system; the reverse calculation can be selected from two-step method, table lookup method and iterative method according to the accuracy requirements, which will be understood by those in the field and will not be elaborated here.

[0094] like Figure 3 As shown, by using the inverse solution of a double prism, the pointing vector of the spatial optical axis can be placed into the rotating prism coordinate system, where each coordinate axis point represents the prism state at a certain moment. The spatial target point planning problem is transformed into a prism motion planning problem that controls the beam direction. Since the cascaded prism inverse solution is a nonlinear mapping, there are multiple sets of equivalent solutions. In the multi-prism coordinate system, this manifests as more than one set of target points on the target solution point map, requiring point elimination and selection during planning. Since the coordinate axes of the multi-prism coordinate system are mutually perpendicular, the distance formulas and definitions in the classical Cartesian coordinate system can be used. The prism motion planning problem can be abstracted into a multi-point traveling salesman problem, and planning selection can be performed using mature algorithms combined with tabu lists.

[0095] S4. In the multi-prism coordinate system, the motion planning of the prism scanner is abstracted as a multi-point traveling salesman problem, planning the motion path of the multi-prism. Simply put, a certain state of the cascaded prism group corresponds to a laser pointing adjustment scheme. Facing multiple unordered targets, a reasonable scheme needs to be designed to complete the scanning of all target points. Therefore, the planned multi-prism motion path is a set of data containing multiple state points, each state point corresponding to a cascaded prism group motion state (the motion angle of each prism). The prism scanner adjusts the prism angles of each prism sequentially according to the planned multi-prism motion path to complete the scanning. For the multi-point traveling salesman problem, existing algorithms can be used for calculation, such as ant colony optimization, genetic algorithms, backtracking, greedy algorithms, and branch and bound methods.

[0096] In terms of overall machine lifespan, the shortest path minimizes losses; however, in terms of search efficiency, the fastest path completes the scan of the predetermined target in the shortest time. There are two different optimization objectives. In actual planning, one objective can be selected based on requirements, or a weighted combination of both objectives can be used.

[0097] To ensure smooth driving and transmission of the prisms, the prisms maintain continuous rotation in the same direction. The cascaded prisms generate 2^n motion modes, where n is the number of prisms. For example, the rotation modes of a rotating double prism are: all clockwise; all counterclockwise; prism 1 clockwise, prism 2 counterclockwise; prism 1 counterclockwise, prism 2 clockwise.

[0098] The optimal path is planned based on different scanning targets, and the prism motion path feature map is drawn according to the same-direction operation rule. The shortest distance, the fastest distance, and the maximum path difference ratio are used as the main basis and evaluation criteria for the overall target scanning path planning. In other implementations, the evaluation criteria can be flexibly adjusted in combination with the optimization target.

[0099] S5. Based on the planned multi-prism motion path, control the prism scanner to scan each target point sequentially.

[0100] S6. Based on the planned multi-prism motion path, the outgoing laser direction is calculated using the iterative vector method, and a predicted target point map is formed in the world coordinate system. The predicted target point map is compared with the target point map obtained from the actual scanning direction in step S5, and the prism scanner motion is corrected.

[0101] This application also provides a prism scanner multi-target detection scanning system. The prism scanner includes a lidar transmitter, a visually guided rangefinder camera, a cascaded prism group, and a support and control system. The visually guided rangefinder camera is mounted off-axis of the lidar transmitter. The cascaded prism group is coaxial and parallel in planar arrangement in front of the lidar transmitter. The scanning system includes:

[0102] The pixel coordinate system module is used to calibrate the camera's intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the pixel coordinate system. The visually guided rangefinder camera is used to photograph multiple targets at a distance, and the targets in the acquired images are identified and marked to obtain a target point map in the pixel coordinate system.

[0103] The multiprism coordinate system module is used to establish a multiprism coordinate system using the coordinate system construction method described above, where the motion angle of the prism is a function of time t.

[0104] The coordinate transformation module is used to establish the inverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism, and realize the transformation from the pixel coordinate system to the multi-prism coordinate system;

[0105] The planning module is used to abstract the motion planning of the prism scanner into a multi-point traveling salesman problem in the multi-prism coordinate system and plan the motion path of the multi-prism.

[0106] The scanning control module is used to control the prism scanner to sequentially scan each target point according to the planned multi-prism motion path.

[0107] The feedback correction module is used to calculate the direction of the emitted laser based on the planned multi-prism motion path using the iterative vector method, form a predicted target point map in the world coordinate system, compare the predicted target point map with the target point map obtained by the actual scanning direction in the scanning control module, and correct the prism scanner motion of the scanning control module.

[0108] The working content of each module in the scanning system is the same as the scanning method described above, and will not be repeated here.

[0109] This invention assigns the prism rotation angles of i cascaded prisms to axes of a Cartesian coordinate system with the same dimension n, where n = i (i ∈ Z+), and a point represents the prism state at a certain moment. By cleverly utilizing the inverse solution of the prisms, the pointing vector of the spatial optical axis is placed into the multi-prism coordinate system, thus enabling direct motion planning of the prism state points. In the multi-prism coordinate system, the motion of the prism scanner can be defined using existing distance definitions in the classical Cartesian coordinate system, and the planning problem can be abstracted into a traveling salesman problem, which can be calculated using existing algorithms.

[0110] The following describes this application using a prism scanner incorporating a rotating double prism, and the application process is as follows: Figure 6 As shown:

[0111] S1. Constructing the camera imaging model: Use Zhang Zhengyou's method to calibrate the camera's intrinsic parameters, obtain the imaging model of the industrial CCD camera, and establish the transformation relationship between the world coordinate system and the camera pixel coordinate system.

[0112] Creating a pixel target point map: A CCD camera is used to capture images of multiple targets at a distance. Based on the images acquired by the camera, a visual recognition algorithm is used to identify and mark 16 feature targets (i.e., identify the objects to be searched). The results are then abstracted and simplified into a target point map of 16 points in a pixel coordinate system, and a map plane is drawn within the pixel coordinate system.

[0113] S2. Establish a multi-prism coordinate system: Establish a classic two-dimensional coordinate system with mutually perpendicular axes. The rotation range of prism 1 corresponds to the X-axis, and the rotation range of prism 2 corresponds to the Y-axis. Each prism rotates independently around a common axis. Let counterclockwise rotation be positive and clockwise rotation be negative. Initially, the thin end of the prism points to the positive X-axis of the world coordinate system, and the rotation angle is a function θ of time t. i (t).

[0114] S21. Define a point in the multiprism coordinate system: each coordinate axis corresponds to the angle of independent motion of different prisms, and a point on a coordinate axis in the coordinate system represents the state of the prism at a certain moment.

[0115] S22. Define the distance between points in the multiprism coordinate system: The total distance traveled by the prism is the algebraic sum of the projections of each target state point onto the coordinate axes, mathematically expressed as the Manhattan distance:

[0116]

[0117] Since the motions of different prisms are relatively independent, the motion time of a prism depends only on the maximum projected distance of the state point on the coordinate axis, mathematically expressed as the Chebyshev distance:

[0118]

[0119] S23. Define the path difference between points in the multiprism coordinate system: The maximum distance the prism moves along a certain coordinate axis is the Chebyshev distance, corresponding to a relative shortest distance. Define the total distance difference of a path along each coordinate axis as the maximum path difference:

[0120]

[0121] For ease of calculation, the maximum road difference ratio is introduced:

[0122]

[0123] S3. Constructing a prism motion inverse calculation model: Using a two-step method, points in the pixel coordinate system are converted into corresponding prism rotation angle values, and a point set is planned. Through Γ2 transformation and coordinate mapping, a multi-objective map is drawn in the multi-prism coordinate system. Since there are two sets of equivalent solutions for the rotating double prism, the points in the 16 pixel coordinate system form a multi-objective solution point map of 32 points in the rotating double prism coordinate system, which serves as the planning target points for the prism scanning motion.

[0124] S4. Planning the motion path of the rotating double prism: In the multi-objective solution point map of the multi-prism coordinate system, there are 16 points to be traversed and their equivalent points. Each path moves only along the coordinate axes and has a certain length. Find a straight path that satisfies the planning objective and passes through each solution point exactly once. In this embodiment, an ant colony algorithm combined with a tabu list is used for path planning.

[0125] S41. In terms of overall machine lifespan, the shortest path results in the least loss; however, in terms of search efficiency, the fastest path can complete the scanning of the given target in the shortest time. This example performs parallel calculations for the two types of scanning targets and plans their corresponding optimal paths respectively.

[0126] S42. To ensure smooth driving and transmission of the prisms, the prisms maintain continuous unidirectional operation. The operating modes of the rotating double prisms are: full forward rotation; full reverse rotation; prism 1 rotates forward, prism 2 rotates reverse; prism 1 rotates reverse, prism 2 rotates forward, marked as Mode++ / Mode-- / Mode+- / Mode-+ respectively.

[0127] S43. Based on different scanning target planning, call the ant colony algorithm combined with the tabu list to calculate and draw the path feature maps of the four prism motion modes in the shortest distance / fastest distance mode.

[0128] The results are as follows Figure 4 As shown, the total distance of each path on the coordinate axis is the sum of the angles rotated by each prism in the prism scanner, which reflects the driving energy and system losses required to complete one scan; the difference in the X and Y directions of movement from the starting point to the ending point of each path is the total path difference of the dual prism operation; the tortuosity of each path shows the smoothness of the scanning path, reflecting the acceleration, deceleration and waiting conditions of the prism.

[0129] Analyze the shortest path, fastest path, and maximum path difference for each path. Based on the scanning target, the preferred path in the prism 1 inverted and prism 2 forward rotation mode is the best path for the shortest distance target; the preferred path in the prism 1 inverted and prism 2 inverted mode is the best path for the fastest distance target.

[0130] S5. Control the prism scanner to scan each target point in sequence: When the host completes the optimal search path planning, it uses position-velocity-time data to synchronously push the motion command into the prism motion controller. The rotating double prism searches for targets in sequence according to the target point order after path optimization, realizing the traversal of all targets.

[0131] S6. Construct a positive pointing model for prism motion: Simultaneously calculate and estimate the laser pointing using the iterative vector refraction method, form a map of the estimated target point in the prism scanner's field of view coordinate system, and provide feedback on the scanning pointing information.

[0132] To highlight the advantages of corner planning for traversing the target trajectory in the prism coordinate system, a comparison is made using a target map planned in the traditional pixel coordinate system. The results are as follows: Figure 5 As shown.

[0133] This embodiment uses a rotating dual-prism scanner and a multi-prism coordinate system for prism scanning path planning. Results show that the multi-prism coordinate system effectively improves traversal search efficiency. Whether in the shortest distance mode or the fastest distance mode, it significantly reduces driving energy and system losses compared to traditional path planning, halving the overall scanning time. The multi-prism coordinate system can provide a complete description of the prism scanner's motion state, offering path planning support for large-scale, long-distance, multi-target scanning and tracking. The planned optimal path reduces overall prism energy consumption and exhibits high operational smoothness and prism utilization.

[0134] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for constructing a coordinate system, characterized in that, Applications include cascaded prism assemblies, including: Establish an n-dimensional coordinate system, where n equals the number of prisms in the cascaded prism group. Define points in the coordinate system as follows: each coordinate axis corresponds to the angle of independent motion of different prisms, and a point on a coordinate axis in the coordinate system represents the state of a prism at a certain moment. Defining the distance between points in a coordinate system: The distance between two points in a coordinate system is expressed as: In the shortest distance mode: Where, θ id θ represents the projection of point a onto coordinate axis i. ib This represents the projection of point b onto coordinate axis i. This represents the distances along all coordinate axes between point a and point b, where points a and b represent time t. a And the next moment t b The prism state, D mk This represents the total distance traveled by the prism when there are l target points in the shortest distance mode. Because the prism rotates simultaneously, in the fastest distance mode: D represents the maximum distance traveled between points a and b. cl This represents the total distance traveled by the prism at the fastest scan speed when there are l target points in the fastest distance mode. Define the path difference between points in the coordinate system: in, D represents the maximum path difference between points a and b; d This represents the maximum total path difference of the prism when there are l target points.

2. The coordinate system construction method according to claim 1, characterized in that, To facilitate the calculation of the maximum reference difference ratio: I d This represents the maximum difference ratio between l target points.

3. A multi-target detection scanning method using a prism scanner, characterized in that, The prism scanner includes a lidar transmitter, a visually guided rangefinder camera, a cascaded prism group, and a support and control system. The visually guided rangefinder camera is positioned off-axis of the lidar transmitter. The cascaded prism group is coaxial and parallel in planar arrangement in front of the lidar transmitter. The scanning method includes the following steps: S1. Calibrate the camera intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the pixel coordinate system; use the visually guided rangefinder camera to take pictures of multiple targets at a distance, identify and mark the targets in the acquired images, and obtain a target point map in the pixel coordinate system. S2. Establish a multi-prism coordinate system using the coordinate system construction method described in any one of claims 1-2, wherein the motion angle of the prism is a function of time t; S3. Establish the inverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism to realize the transformation from the pixel coordinate system to the multi-prism coordinate system; S4. In the multiprism coordinate system, the motion planning of the prism scanner is abstracted into a multi-point traveling salesman problem, and the motion path of the multiprism is planned. S5. Based on the planned multi-prism motion path, control the prism scanner to scan each target point sequentially.

4. The multi-target detection scanning method of a prism scanner according to claim 3, characterized in that, Also includes: S6. Based on the planned multi-prism motion path, the outgoing laser direction is calculated using the iterative vector method, and a predicted target point map is formed in the world coordinate system. The predicted target point map is compared with the target point map obtained from the actual scanning direction in step S5, and the prism scanner motion is corrected.

5. The multi-target detection scanning method of a prism scanner according to claim 3, characterized in that, In step S1, the camera intrinsic parameters are calibrated using the Zhang Zhengyou method.

6. The multi-target detection scanning method of a prism scanner according to claim 3, characterized in that, In step S4, the optimization objective is to find the shortest path, the fastest path, or a weighted combination of the two.

7. The multi-target detection scanning method of a prism scanner according to claim 3, characterized in that, In step S4, the cascaded prism group generates 2n motion modes, where n is the number of prisms.

8. The multi-target detection scanning method of a prism scanner according to claim 3, characterized in that, In step S4, for the obtained multi-prism motion path, a prism motion path feature map is drawn according to the same-direction operation rule, and the shortest distance, the fastest distance, and the maximum path difference ratio are used as the evaluation criteria for the path.

9. A prism scanner multi-target detection scanning system, characterized in that, The prism scanner includes a lidar transmitter, a visually guided rangefinder camera, a cascaded prism group, and a support control system. The visually guided rangefinder camera is mounted off-axis of the lidar transmitter. The cascaded prism group is coaxial and parallel in plane, positioned directly in front of the lidar transmitter. The scanning system includes: The pixel coordinate system module is used to calibrate the camera's intrinsic parameters, obtain the imaging model of the visually guided rangefinder camera, and establish the transformation relationship between the world coordinate system and the pixel coordinate system. The visually guided rangefinder camera is used to photograph multiple targets at a distance, and the targets in the acquired images are identified and marked to obtain a target point map in the pixel coordinate system. A multiprism coordinate system module is used to establish a multiprism coordinate system using the coordinate system construction method described in any one of claims 1-2, wherein the motion angle of the prism is a function of time t; The coordinate transformation module is used to establish the inverse calculation relationship between the optical axis pointing angle and the motion values ​​of each prism, and realize the transformation from the pixel coordinate system to the multi-prism coordinate system; The planning module is used to abstract the motion planning of the prism scanner into a multi-point traveling salesman problem in the multi-prism coordinate system and plan the motion path of the multi-prism. The scanning control module is used to control the prism scanner to sequentially scan each target point according to the planned multi-prism motion path.

10. A prism scanner multi-target detection scanning system according to claim 9, characterized in that, Also includes: The feedback correction module is used to calculate the direction of the emitted laser based on the planned multi-prism motion path using the iterative vector method, form a predicted target point map in the world coordinate system, compare the predicted target point map with the target point map obtained by the actual scanning direction in the scanning control module, and correct the prism scanner motion of the scanning control module.

Images