Anti-collision method, device and equipment of six-degree-of-freedom treatment bed, medium and program product

Abstract

The invention provides an anti-collision method of a six-degree-of-freedom treatment bed, and relates to the field of intelligent anti-collision. The anti-collision method comprises the following steps: acquiring point cloud data of a static environment in a radiotherapy room, and generating a static point cloud model after performing point cloud reconstruction processing on the point cloud data; comparing the movement speed of the dynamic obstacle with a preset speed threshold value, and collecting point cloud data of the dynamic obstacle in a classified manner; fusing the point cloud data of the first dynamic obstacle with the static point cloud model to obtain a background point cloud model; acquiring a local dynamic point cloud model of the six-degree-of-freedom treatment bed, and respectively calculating a minimum static distance between the local dynamic point cloud model and the background point cloud model and a minimum dynamic distance between the local dynamic point cloud model and the point cloud data of the second dynamic obstacle; and controlling the movement of the six-degree-of-freedom treatment bed according to the minimum static distance, the minimum dynamic distance and the safe distance threshold. The invention further provides an anti-collision device and equipment for the six-degree-of-freedom treatment bed, a storage medium and a program product.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of intelligent anti-collision, and more particularly to an anti-collision method, device, equipment, medium and program product for a six-degree-of-freedom treatment bed based on point cloud. BACKGROUND

[0002] In a radiotherapy process, the automatic operation of large high-precision devices such as a treatment bed and an accelerator gantry is a key to improving treatment efficiency and accuracy.

[0003] The existing mainstream technology mainly relies on manual measurement and three-dimensional modeling. A technician needs to spend a lot of time to accurately manually measure the walls, cabinets and all devices in the treatment room, and manually build a scene model in CAD drawing software or three-dimensional modeling software.

[0004] For devices with complex motion joints, such as a six-degree-of-freedom treatment bed, a method of creating a maximum motion envelope is usually used for simplified processing, but this method cannot adjust the anti-collision threshold or detection accuracy in real time and dynamically according to the actual motion state of the device or changes in the environment during anti-collision warning. In addition, when the motion range of the device or the layout of the environment changes, the bounding box model often needs to be manually reconstructed, and the geometric simplification of the bounding box itself sacrifices the detection accuracy, making it difficult to accurately capture the subtle spatial relationship between the complex shape of the device and the obstacles. SUMMARY

[0005] In view of the above problems, the present disclosure provides an anti-collision method, device, equipment, medium and program product for a six-degree-of-freedom treatment bed, which can meet the real-time safety protection needs of the six-degree-of-freedom treatment bed under high-speed and high-precision motion in a radiotherapy room scene.

[0006] According to a first aspect of the present disclosure, an anti-collision method for a six-degree-of-freedom treatment bed based on point cloud is provided, comprising: acquiring point cloud data of a static environment in a radiotherapy room, generating a static point cloud model after point cloud reconstruction processing of the point cloud data; detecting the motion speed of a plurality of dynamic obstacles in the static environment, and classifying and collecting the point cloud data of the dynamic obstacles according to the comparison result of the motion speed and a preset speed threshold; fusing the point cloud data of a first dynamic obstacle in the plurality of dynamic obstacles with the static point cloud model to obtain a background point cloud model, the motion speed of the first dynamic obstacle being lower than the preset speed threshold; acquiring a local dynamic point cloud model of the six-degree-of-freedom treatment bed, calculating the minimum static distance between the local dynamic point cloud model and the background point cloud model, and the minimum dynamic distance between the local dynamic point cloud model and the point cloud data of a second dynamic obstacle in the plurality of dynamic obstacles, the motion speed of the second dynamic obstacle being not lower than the preset speed threshold; and controlling the motion state of the six-degree-of-freedom treatment bed according to the comparison result of the smaller value of the minimum static distance and the minimum dynamic distance and a safety distance threshold.

[0007] According to an embodiment of the present disclosure, when the motion speed of any one of the plurality of dynamic obstacles is lower than a preset motion threshold, the corresponding dynamic obstacle is determined as a first dynamic obstacle, and the point cloud data of the first dynamic obstacle is obtained through low-frequency scanning; when the motion speed of any one of the plurality of dynamic obstacles is not lower than the preset motion threshold, the corresponding dynamic obstacle is determined as a second dynamic obstacle, and the point cloud data of the second dynamic obstacle is obtained through a sensing device.

[0008] According to an embodiment of the present disclosure, the point cloud data of the first dynamic obstacle is registered to the coordinate system of the static point cloud model through a rigid body transformation matrix; an overlapping region of the point cloud data of the first dynamic obstacle and the static point cloud model is identified, when the overlapping region is identified, the static point cloud model is reconstructed according to the point cloud data in the overlapping region to obtain a background point cloud model; and when the overlapping region cannot be identified, the point cloud data of the first dynamic obstacle is superimposed into the static point cloud model to obtain the background point cloud model.

[0009] According to an embodiment of the present disclosure, the point cloud data in the overlapping region and the static point cloud model are weighted and fused to obtain fused point cloud data; and the fused point cloud data is reconstructed to obtain the background point cloud model.

[0010] According to an embodiment of the present disclosure, sub-point cloud data of n treatment bed sub-components in a six-degree-of-freedom treatment bed in a local coordinate system is collected in real time through a motion sensor, the n treatment bed sub-components are connected according to a motion chain relationship, and the parent index of an nth treatment bed sub-component is n-1; for the sub-point cloud data of the nth treatment bed sub-component, the following operations are performed: a rotation amount and a translation amount of the nth treatment bed sub-component relative to a parent body corresponding to the treatment bed sub-component are obtained through the motion sensor; a pose transformation matrix is determined according to the rotation amount and the translation amount; and motion transformation chain calculation is performed on the sub-point cloud data based on a coordinate system transformation matrix and the pose transformation matrix to obtain point cloud data in a coordinate system in a target space, wherein the coordinate system transformation matrix is determined according to the position relationship of the six-degree-of-freedom treatment bed relative to a coordinate system in a radiotherapy room; and when the index n of the treatment bed sub-component is incremented to N, N corresponding point cloud data are taken as the point cloud data of the six-degree-of-freedom treatment bed.

[0011] According to an embodiment of the present disclosure, a spatial index structure of a background point cloud model and a spatial index structure of a second dynamic obstacle are constructed; point cloud data in a local dynamic point cloud model is taken as a query point set, the query point set includes m pieces of point cloud data, m = 1, 2, 3, …, M, and the following operations are performed on the m pieces of point cloud data in the query point set: the nearest neighbor point of the mth piece of point cloud data in the background point cloud model is determined through the spatial index structure of the background point cloud model; the Euclidean distance between the mth piece of point cloud data and the nearest neighbor point is calculated; when the query point set index m is incremented to M, the minimum value of the M Euclidean distances is taken as a minimum static distance; the nearest neighbor point of the mth piece of point cloud data in the point cloud data of the second dynamic obstacle is determined through the spatial index structure of the second dynamic obstacle; the Euclidean distance between the mth piece of point cloud data and the corresponding nearest neighbor point is calculated; when the query point set index m is incremented to M, the minimum value of the M Euclidean distances is taken as a minimum dynamic distance.

[0012] According to an embodiment of the present disclosure, the smaller value of the minimum static distance and the minimum dynamic distance is taken as a predicted distance value; when the predicted distance is less than or equal to a safety distance threshold value, a first-level control instruction is generated to control the six-degree-of-freedom treatment bed to immediately terminate movement; when the predicted distance is greater than the safety distance threshold value, a second-level control instruction is generated to control the six-degree-of-freedom treatment bed to immediately decelerate movement or issue a warning signal.

[0013] Another aspect of the present disclosure discloses a point cloud-based anti-collision device of a six-degree-of-freedom treatment bed, the device comprising: a cloud model generation module, which acquires point cloud data of a static environment in a radiotherapy room, and generates a static point cloud model after point cloud reconstruction processing of the point cloud data; a classification acquisition module, which detects the motion speed of a plurality of dynamic obstacles in the static environment, and classifies and acquires the point cloud data of the dynamic obstacles according to the comparison result of the motion speed and a preset speed threshold value; a model update module, which fuses the point cloud data of a first dynamic obstacle in the plurality of dynamic obstacles with the static point cloud model to obtain a fused background point cloud model, the motion speed of the first dynamic obstacle being lower than the preset speed threshold value; a distance calculation module, which acquires a local dynamic point cloud model of the six-degree-of-freedom treatment bed, and respectively calculates a minimum static distance between the local dynamic point cloud model and the background point cloud model, and a minimum dynamic distance between the local dynamic point cloud model and the point cloud data of a second dynamic obstacle in the plurality of dynamic obstacles, the motion speed of the second dynamic obstacle being not lower than the preset speed threshold value; and an operation control module, which controls the motion state of the six-degree-of-freedom treatment bed according to the comparison result of the smaller value of the minimum static distance and the minimum dynamic distance and a safety distance threshold value.

[0014] The point cloud-based anti-collision method, device, equipment, medium and program product of the six-degree-of-freedom treatment bed provided by the embodiments of the present disclosure have at least the following beneficial effects:

[0015] (1) By distinguishing obstacles in the treatment room scene, the point data corresponding to different categories of obstacles are processed differently;

[0016] (2) Based on high-density real object surface point cloud, distance calculation can accurately perceive objects, ensure millimeter-level collision detection accuracy, and significantly improve the safety of the treatment process;

[0017] (3) Real-time point cloud data is used to dynamically construct and update scene point cloud models, and distance calculation is combined to optimize the efficiency and effect of collision detection, which significantly improves the realism, real-time performance and flexibility of the anti-collision method. Attached Figure Description

[0018] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0019] Figure 1 A flowchart illustrating a six-degree-of-freedom treatment bed anti-collision method according to an embodiment of the present disclosure is shown schematically.

[0020] Figure 2 A schematic diagram illustrating the structure of a six-degree-of-freedom treatment bed anti-collision device according to an embodiment of the present disclosure is shown; and

[0021] Figure 3 A block diagram schematically illustrates an electronic device suitable for implementing a six-degree-of-freedom treatment bed anti-collision method according to an embodiment of the present disclosure. Detailed Implementation

[0022] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0023] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0024] All terms used herein, including technical and scientific terms, have the meanings commonly understood by one of ordinary skill in the art unless otherwise defined. It should be further noted that the use of terms such as first and second should be understood in the context of the specification as a whole, and not in a manner that is overly literal or dogmatic.

[0025] In the case of using expressions similar to "at least one of A, B, and C, etc.", it should be generally interpreted that the meaning of the expression is at least one of A, B, and C (for example, "a system having at least one of A, B, and C" should include but not be limited to a system having A alone, a system having B alone, a system having C alone, a system having both A and B, a system having both A and C, a system having both B and C, and / or a system having A, B, and C, etc.).

[0026] Figure 1 A flowchart of a collision avoidance method of a six-degree-of-freedom treatment bed according to an embodiment of the disclosure is schematically shown.

[0027] As shown in Figure 1 The collision avoidance method of the six-degree-of-freedom treatment bed based on point cloud of the embodiment includes operations S110-S150.

[0028] In operation S110, point cloud data of a static environment in a radiotherapy room is acquired, and a static point cloud model is generated after point cloud reconstruction processing is performed on the point cloud data.

[0029] In the embodiment of the disclosure, the static environment includes walls and equipment that cannot be moved in the scene of the hospital radiotherapy room, such as fixed cabinets.

[0030] A large number of three-dimensional coordinate points of surfaces of walls, ceilings, floors, and all fixed equipment in the radiotherapy room can be acquired at one time by a high-precision laser scanner as point cloud data of the static environment. Then, the topological structure of the three-dimensional coordinate points is mapped and reconstructed to obtain a high-density and high-precision static point cloud model, and the static background point cloud model is fixed in a world coordinate system. Optionally, the base point of the world coordinate system is set as a corner of the radiotherapy room or a center point of a support plate of the six-degree-of-freedom treatment bed.

[0031] In operation S120, the motion speed of a plurality of dynamic obstacles in the static environment is detected, and the point cloud data of the dynamic obstacles is classified and collected according to a comparison result of the motion speed and a preset speed threshold.

[0032] In the embodiments of the present disclosure, the dynamic obstacles in the static environment can be multiple, and the dynamic obstacles include a first dynamic obstacle. By detecting the movement speed of multiple obstacles in the radiotherapy room, a preset speed threshold is set, the speed of the multiple obstacles is compared with the preset speed threshold, when the movement speed of the dynamic obstacle is higher than the preset speed threshold, the point cloud data is collected in real time by the sensor, and when the movement speed of the dynamic obstacle is lower than the preset speed threshold, the point cloud data is collected by low-frequency scanning.

[0033] It should be noted that the preset speed threshold is optional, and different preset speed thresholds can be set according to different running times of the radiotherapy room.

[0034] In operation S130, the point cloud data of the first dynamic obstacle in the multiple dynamic obstacles is fused with the static point cloud model to obtain a fused background point cloud model, and the movement speed of the first dynamic obstacle is lower than the preset speed threshold.

[0035] In the embodiments of the present disclosure, the first dynamic obstacle includes a series of non-fixed devices or furniture such as chairs, cabinets and the like in the radiotherapy room. The point cloud data representing the contour of the first dynamic obstacle is fused with the static point cloud model representing the contour of the wall and the fixed device, so as to obtain a fused background point cloud model, which can completely reflect the internal structure background point cloud model of the radiotherapy room.

[0036] It should be noted that the first dynamic obstacle is a non-fixed obstacle, which can change position or state due to external force. Optionally, the actual position information of the first dynamic obstacle is input or the point cloud data of the first dynamic obstacle is collected by a low-frequency sensor to obtain the point cloud data of the first dynamic obstacle. The real-time position information is dynamically placed to the static point cloud model in the world coordinate system, and a relatively stable background point cloud model is formed by fusion, which represents the overall structure of the objects fixed or infrequently moved in the radiotherapy room.

[0037] In operation S140, a local dynamic point cloud model of a six-degree-of-freedom treatment bed is obtained, the minimum static distance between the local dynamic point cloud model and the background point cloud model is calculated, and the minimum dynamic distance between the local dynamic point cloud model and the point cloud data of a second dynamic obstacle in the multiple dynamic obstacles is calculated, and the movement speed of the second dynamic obstacle is not lower than the preset speed threshold.

[0038] In the embodiments of the present disclosure, the second dynamic obstacle includes medical staff and devices that can move quickly, such as infusion stands and wheelchairs. The local dynamic point cloud model of the six-degree-of-freedom treatment bed includes multiple point cloud data of sub-components of the six-degree-of-freedom treatment bed, and the point cloud data represents the position information of various sub-devices of the six-degree-of-freedom treatment bed.

[0039] For each of the plurality of point cloud data in the local dynamic point cloud model, the point cloud data is taken as a target point cloud data, and the plurality of point cloud data in the background point cloud model is traversed to calculate the distance between the target point cloud data and each point cloud data in the background point cloud model. In this way, the distance between each target point cloud data in the local dynamic point cloud model and each point cloud data in the background point cloud model can be calculated. Then, the minimum distance between the plurality of target point cloud data and each point cloud data in the background point cloud model is taken as the minimum static distance between the local dynamic point cloud model of the six-degree-of-freedom treatment bed and the static obstacle in the radiotherapy room.

[0040] For example, the local dynamic point cloud model includes 5 target point cloud data, and the background point cloud model includes 6 point cloud data. Based on the above method, the distance between each target point cloud data and each point cloud data in the background point cloud model can be calculated, and 30 distance values are obtained. Then, the minimum distance can be determined from the 30 distance values, and the minimum distance is taken as the minimum static distance between the local dynamic point cloud model of the six-degree-of-freedom treatment bed and the static obstacle in the radiotherapy room.

[0041] For each of the plurality of point cloud data in the local dynamic point cloud model, the point cloud data is taken as a target point cloud data, and the plurality of point cloud data in the background point cloud model is traversed to calculate the distance between the target point cloud data and each point cloud data in the background point cloud model. In this way, the distance between each target point cloud data in the local dynamic point cloud model and each point cloud data in the background point cloud model can be calculated. Then, the minimum distance between the plurality of target point cloud data and each point cloud data in the background point cloud model is taken as the minimum static distance between the local dynamic point cloud model of the six-degree-of-freedom treatment bed and the static obstacle in the radiotherapy room.

[0042] For example, the local dynamic point cloud model includes 5 target point cloud data, and the background point cloud model includes 6 point cloud data. Based on the above method, the distance between each target point cloud data and each point cloud data in the background point cloud model can be calculated, and 30 distance values are obtained. Then, the minimum distance can be determined from the 30 distance values, and the minimum distance is taken as the minimum static distance between the local dynamic point cloud model of the six-degree-of-freedom treatment bed and the static obstacle in the radiotherapy room.

[0043] In operation S150, the motion state of the six-degree-of-freedom treatment bed is controlled according to the comparison result of the smaller value of the minimum static distance and the minimum dynamic distance and the safety distance threshold.

[0044] In the embodiments of the present disclosure, the smaller value is selected from the minimum static distance and the minimum dynamic distance as the early warning reference distance, the early warning reference distance is compared with the safety distance threshold, and the motion state of the six-degree-of-freedom treatment bed is controlled according to the comparison result, so that the six-degree-of-freedom treatment bed avoids the dynamic and static obstacles. In this way, distance calculation is performed based on high-density real object surface point cloud, the oversimplified envelope body model is abandoned, the subtle spatial relationship of the object can be accurately perceived, the millimeter-level collision detection accuracy is ensured, and the safety of the treatment process is significantly improved.

[0045] The operation process of operation S110 to operation S150 will be described in detail below.

[0046] In operation S110, the static environment (such as walls, building structures, and static objects in the radiotherapy room) of the radiotherapy room is scanned by a laser scanner to obtain point cloud data spatial structure point data. Then, a modeling tool is combined to construct a static point cloud model with accurate proportions. The static point cloud model is a three-dimensional model, and is imported into a graphics engine to generate a virtual space corresponding to the real physical scene 1:1.

[0047] Operation S120 can further include operation S121 to operation S122.

[0048] In operation S121, when the motion speed of any dynamic obstacle in the plurality of dynamic obstacles is lower than a preset motion threshold, the corresponding dynamic obstacle is determined as a first dynamic obstacle, and the point cloud data of the first dynamic obstacle is obtained by low-frequency scanning.

[0049] In the embodiments of the present disclosure, the plurality of dynamic obstacles in the radiotherapy room are continuously measured, when the motion speed of the dynamic obstacle is lower than the preset threshold, the corresponding dynamic obstacle is classified as a first dynamic obstacle, and the point cloud data of the first dynamic obstacle is obtained by low-frequency scanning.

[0050] It should be noted that the plurality of dynamic obstacles in the radiotherapy room can also be continuously monitored for motion state, when it is detected that the motion state of the dynamic obstacle is intermittent movement, the corresponding dynamic obstacle is classified as a first dynamic obstacle, and the point cloud data of the first dynamic obstacle is obtained by low-frequency scanning.

[0051] In operation S122, when the motion speed of any dynamic obstacle in the plurality of dynamic obstacles is not lower than the preset motion threshold, the corresponding dynamic obstacle is determined as a second dynamic obstacle, and the point cloud data of the second dynamic obstacle is obtained by a sensing device.

[0052] It should be noted that, by continuously monitoring the motion state of the plurality of dynamic obstacles in the radiotherapy room, when it is detected that the motion state of the dynamic obstacle is continuous movement, the corresponding dynamic obstacle is classified as a second dynamic obstacle, and the point cloud data of the second dynamic obstacle is obtained in real time through the sensor, and by differentiating the point cloud data acquisition of the dynamic obstacles in the radiotherapy room, the amount of data that needs to be processed subsequently is reduced, and the calculation efficiency is improved.

[0053] The operation S130 can further include operations S131-S134.

[0054] In operation S131, the point cloud data of the first dynamic obstacle is registered to the coordinate system of the static point cloud model through the rigid transformation matrix.

[0055] In the embodiments of the present disclosure, the point cloud data of the first dynamic obstacle is converted into the coordinate system of the static point cloud model according to the rigid transformation matrix T, and the matrix form of T is:

[0056]

[0057] wherein R∈R {3×3} represents a rotation matrix, and t∈R {3} represents a translation vector, and R and t can be obtained by converting the point cloud data obtained by the sensor.

[0058] In operation S132, the overlapping area of the point cloud data of the first dynamic obstacle and the static point cloud model is identified, and when the overlapping area is identified, the static point cloud model is reconstructed according to the point cloud data in the overlapping area to obtain a background point cloud model.

[0059] In the embodiments of the present disclosure, the point cloud data of the first dynamic obstacle and the point cloud data in the static point cloud model are searched and compared through spatial hashing or KD tree, and the existence or nonexistence of the overlapping area of the point cloud data of the first dynamic obstacle and the static model is identified through the comparison result. When it is identified that the overlapping area exists, the static point cloud model is reconstructed based on the point cloud data in the overlapping area to obtain the reconstructed background point cloud model.

[0060] In operation S132, the static point cloud model is reconstructed according to the point cloud data in the overlapping area to obtain the background point cloud model, including operations S1321-S1322.

[0061] In operation S1321, the point cloud data in the overlapping area and the static point cloud model are weighted and fused to obtain the fused point cloud data.

[0062] In operation S1322, the fused point cloud data is reconstructed to obtain the reconstructed background point cloud model.

[0063] In the embodiments of the present disclosure, when it is identified that there is an overlapping region, the point cloud data in the overlapping region is linearly interpolated by a weight coefficient a:

[0064] P_fused = a * P_dynamic + (1-a) * P_static

[0065] wherein P_fused represents the reconstructed background point cloud model, P_dynamic represents the point cloud data in the overlapping region, the weight coefficient a is determined by the dynamic obstacle movement speed or the point cloud confidence, for example, a low-speed dynamic obstacle such as a moving chair, a tends to 0.5, and a high-confidence static point such as a wall surface, a tends to 0, and the static point cloud model is preferentially retained.

[0066] In operation S133, when the overlapping region cannot be identified, the point cloud data of the first dynamic obstacle is superimposed into the static point cloud model to obtain the background point cloud model.

[0067] In the embodiments of the present disclosure, when the overlapping region cannot be identified, the point cloud data in the non-overlapping region is retained, and the point cloud data of the first dynamic obstacle is directly superimposed into the static point cloud model to obtain the background point cloud model.

[0068] It should be noted that after obtaining the fused background point cloud model, a spatial index data structure is constructed for the background point cloud model, for example, KD-Tree or spatial hashing (Spatial Hashing). Since the background point cloud model is relatively stable, only local reconstruction index updating occurs at the position of the first dynamic obstacle, and the construction of the background point cloud model spatial index data can greatly speed up the subsequent query process of the point cloud data in the background point cloud model.

[0069] The above operation S140 can further include operations S141-S142.

[0070] In operation S141, a spatial index structure of the background point cloud model is constructed, and the point cloud data in the local dynamic point cloud model is taken as a query point set, the query point set includes m point cloud data, m = 1, 2, 3, …, M, and the following operations are performed for the mth point cloud data in the query point set: the nearest neighbor point of the mth point cloud data in the background point cloud model is determined through the spatial index structure of the background point cloud model; the Euclidean distance between the mth point cloud data and the nearest neighbor point is calculated; and when the query point set index m is incremented to M, the minimum value of the M Euclidean distances is taken as the minimum static distance.

[0071] In the embodiment of the present disclosure, all point cloud data in the local dynamic point cloud model of the six-degree-of-freedom treatment bed is taken as a query object, the nearest neighbor point in the background point cloud model is calculated by taking each point cloud data in the local dynamic point cloud model as a target point cloud data through the constructed background point cloud model KD-Tree or equivalent efficient index, the nearest neighbor distance between the target point cloud data and the nearest neighbor point in the background point cloud model is obtained, the minimum value d_bg of the nearest neighbor distance is determined by traversing all the nearest neighbor distances, and the minimum value d_bg represents the minimum distance between the treatment bed and the static obstacle and the first dynamic obstacle.

[0072] In operation S142, a spatial index structure of the second dynamic obstacle is constructed, the nearest neighbor point of the mth point cloud data in the point cloud data of the second dynamic obstacle is determined through the spatial index structure of the second dynamic obstacle, the Euclidean distance between the mth point cloud data and the corresponding nearest neighbor point is calculated, and when the query point set index m is incremented by M, the minimum value in the M Euclidean distances is taken as the minimum dynamic distance.

[0073] In the embodiment of the present disclosure, all point cloud data in the local dynamic point cloud model of the six-degree-of-freedom treatment bed is taken as a query object, the nearest neighbor point in the background point cloud model is calculated by taking each point cloud data in the local dynamic point cloud model as a target point cloud data through the constructed background point cloud model KD-Tree or equivalent efficient index, the nearest neighbor distance between the target point cloud data and the nearest neighbor point in the background point cloud model is obtained, the minimum value d_bg of the nearest neighbor distance is determined by traversing all the nearest neighbor distances, and the minimum value d_bg represents the minimum distance between the treatment bed and the static obstacle and the first dynamic obstacle.

[0074] It should be noted that the global minimum distance value d_min is the minimum value of d_bg and d_dyn, and the global minimum distance d_min = min(d_bg, d_dyn). The global minimum distance d_min represents the shortest distance between the six-degree-of-freedom treatment bed and any type of obstacle in the environment at the current moment, and the KD-Tree or equivalent index structure optimizes the average time complexity of the nearest neighbor search from O(N) to O(logN), ensuring that the complex query calculation can be completed within the strict real-time control period.

[0075] It should be noted that before S140, the sub-point cloud data of n treatment bed sub-components in the local coordinate system of the six-degree-of-freedom treatment bed is collected in real time through a motion sensor, the n treatment bed sub-components are connected according to a motion chain relationship, and the parent index of the nth treatment bed sub-component is n-1.

[0076] For the sub-point cloud data of the ith treatment bed sub-component, the following operations are performed:

[0077] The rotation amount and the translation amount of the nth treatment bed sub-component relative to the corresponding parent body of the treatment bed sub-component are acquired by the motion sensor; a pose transformation matrix is determined according to the rotation amount and the translation amount; the motion transformation chain calculation is performed on the sub-point cloud data based on the coordinate system transformation matrix and the pose transformation matrix, to obtain the point cloud data under the coordinate system in the target space, wherein the coordinate system transformation matrix is determined according to the position relationship of the six-degree-of-freedom treatment bed relative to the position relationship in the radiotherapy room; when the index n of the treatment bed sub-component itself is incremented to N, the N corresponding point cloud data are taken as the point cloud data of the six-degree-of-freedom treatment bed.

[0078] In the embodiments of the present disclosure, the six-degree-of-freedom treatment bed is composed of i rigid sub-components, and the rigid sub-components are linked by a motion chain relationship, wherein i = 1, 2, 3 .

[0079] It is assumed that the coordinate system in the radiotherapy room is W, and the point cloud data of the ith sub-component in the local coordinate system L i of the sub-component itself is:

[0080]

[0081] wherein, is a three-dimensional column vector [x, y, z] T .

[0082] At time t, the pose transformation of the ith component relative to its parent component (i-1) obtained from the sensor can be represented by a 4x4 homogeneous transformation matrix .

[0083]

[0084] wherein, is a 3x3 rotation matrix, is a 3x1 translation vector, wherein the rotation matrix and the translation vector are determined in real time by the sensor data of the rigid sub-component.

[0085] For any point on the rigid sub-component i, the coordinates of the point in the radiotherapy room coordinate system W can be calculated by the motion chain relationship formula from the local coordinate of the rigid sub-component i to the radiotherapy room coordinate system:

[0086]

[0087] wherein, is the transformation matrix of the device base relative to the radiotherapy room coordinate system, and optionally, the slide sleeve base of the six-degree-of-freedom treatment bed is set as the base point.

[0088] The world coordinate point cloud of the rigid subcomponent i at the time t is obtained by transforming all the local data points of the rigid subcomponent i The complete dynamic point cloud of the six-degree-of-freedom treatment bed is the union of the point cloud data of all the subcomponents thereof in the coordinate system of the radiotherapy room:

[0089]

[0090] It should be noted that the six-degree-of-freedom treatment bed is composed of a plurality of subcomponents with independent degrees of freedom, and a position sensor is arranged on each rotating shaft or sliding rail to obtain the attitude change in real time, and the point data of each component is extracted to form a vertex set based on the local coordinate system from a plurality of point data sets, and according to the current sensor data and the preset kinematic model, the transformation matrix of each subcomponent in the world coordinate system is calculated, and then the local point set is mapped to the actual space point set in the world coordinate system, and the dynamic structure point set of the current device as a whole is combined in real time.

[0091] The above operation S150 can further include operations S151-S152.

[0092] In operation S151, the smaller value of the minimum static distance and the minimum dynamic distance is taken as the predicted distance value; when the predicted distance is less than or equal to the safety distance threshold, a first-level control instruction is generated to control the six-degree-of-freedom treatment bed to immediately terminate the movement.

[0093] In the embodiments of the present disclosure, when the global minimum distance d_min is less than the preset safety distance threshold, a six-degree-of-freedom treatment bed stop movement instruction is triggered immediately to prevent collision.

[0094] In operation S152, when the predicted distance is greater than the safety distance threshold, a second-level control instruction is generated to control the six-degree-of-freedom treatment bed to immediately decelerate the movement or issue a warning signal.

[0095] In the embodiments of the present disclosure, when the global minimum distance d_min is greater than the safety distance threshold, a warning signal such as an audible and visual alarm can be triggered, or a second-level control instruction is generated to make the six-degree-of-freedom treatment bed decelerate the movement, providing a buffer time for operator intervention or system automatic adjustment.

[0096] It should be noted that the customized distance information can also be output, such as the minimum distance d_bg of the local dynamic point cloud model and the second dynamic obstacle, the minimum distance d_dyn of the six-degree-of-freedom treatment bed and the static obstacle and the second dynamic obstacle, or the nearest distance of a specific area, which is used for the control strategy and does not depend on the specific path planning algorithm implementation, but as an independent and functional complete environmental distance perception method, the functional decoupling is realized in the method design, and the system compatibility and integrability are achieved.

[0097] Based on the method provided in the embodiments of the present disclosure, the following beneficial effects are achieved:

[0098] First, distance calculation is performed based on high-density real object surface point cloud, and an oversimplified envelope model is abandoned, so that the object can be accurately perceived, millimeter-level collision detection accuracy is ensured, and the safety of the treatment process is significantly improved.

[0099] Second, the time-consuming and laborious full manual measurement and manual modeling link in the traditional scheme is replaced by automatic or semi-automatic laser scanning technology and programmed point cloud processing process, so that the deployment and calibration time of the system is greatly shortened, and the implementation cost is significantly reduced.

[0100] Third, the posture change of each component point cloud is driven in real time by the sensor, so that the virtual model and the physical device are synchronized in microseconds, and the KD-Tree space index and the optimized nearest neighbor search algorithm are combined to ensure that the real-time distance calculation of the complex scene is completed within a high-speed motion control period, for example, 10ms, and the accurate minimum distance value is output, and multiple safety threshold values including warning, deceleration, and stopping are supported, so that the collision risk is managed in stages, and valuable advance intervention time is provided for the operator or the control system.

[0101] Fourth, without complex code-level modification of the core collision detection algorithm, the system can automatically or quickly adapt to the new environment, and the maintenance is very convenient.

[0102] Based on the six-degree-of-freedom treatment bed anti-collision method described above, the embodiments of the present disclosure further provide a six-degree-of-freedom treatment bed anti-collision device. The following will be described in detail Figure 2 The device.

[0103] Figure 2 The structure block diagram of the point cloud-based six-degree-of-freedom treatment bed anti-collision device according to the embodiments of the present disclosure is schematically shown.

[0104] As Figure 2 shown, the six-degree-of-freedom treatment bed anti-collision device 200 of the embodiments includes a cloud model generation module 210, a classification collection module 220, a cloud model update module 230, a distance calculation and detection module 240, and an operation control module 250.

[0105] The cloud model generation module 210 is configured to obtain point cloud data of a static environment in a radiotherapy room, and generate a static point cloud model after point cloud reconstruction processing of the point cloud data. In an embodiment, the cloud model generation module 210 can be configured to perform the operation S210 described above, and details are not repeated here.

[0106] The classification acquisition module 220 is configured to detect the motion speed of the dynamic obstacle in the static environment, and according to the comparison result of the motion speed and the preset speed threshold, classify and acquire the point cloud data of the dynamic obstacle. The classification acquisition module 220 can be configured to perform the operation S220 described above, and details are not repeated here.

[0107] The cloud model updating module 230 is configured to fuse the point cloud data of the first dynamic obstacle with the static point cloud model to obtain a fused background point cloud model, and the motion speed of the first dynamic obstacle is lower than the preset speed threshold. The cloud model updating module 230 can be configured to perform the operation S230 described above, and details are not repeated here.

[0108] The distance calculation module 240 is configured to obtain a local dynamic point cloud model of the six-degree-of-freedom treatment bed, and calculate the minimum static distance between the local dynamic point cloud model and the background point cloud model and the minimum dynamic distance between the local dynamic point cloud model and the point cloud data of the second dynamic obstacle, respectively, and the motion speed of the second dynamic obstacle is higher than or equal to the preset speed threshold. The distance calculation module 240 can be configured to perform the operation S240 described above, and details are not repeated here.

[0109] The operation control module 250 is configured to control the motion of the six-degree-of-freedom treatment bed according to the comparison result of the smaller value of the minimum static distance and the minimum dynamic distance and the safety distance threshold. The control module 250 can be configured to perform the operation S250 described above, and details are not repeated here.

[0110] According to embodiments of the present disclosure, any multiple of the cloud model generating module 210, the classification collecting module 220, the cloud model updating module 230, the distance calculating module 240 and the operation control module 250 can be combined in one module, or any one of them can be split into multiple modules. Alternatively, at least part of the function of one or more of these modules can be combined with at least part of the function of other modules and implemented in one module. According to embodiments of the present disclosure, at least one of the cloud model generating module 210, the classification collecting module 220, the cloud model updating module 230, the distance calculating module 240 and the operation control module 250 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on board, a system on package, an application specific integrated circuit (ASIC), or any other reasonable way of integrating or packaging a circuit, etc. hardware or firmware, or in any one of the three implementation ways of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the cloud model generating module 210, the classification collecting module 220, the cloud model updating module 230, the distance calculating module 240 and the operation control module 250 can be at least partially implemented as a computer program module which, when executed, can perform the corresponding functions.

[0111] Figure 3 A block diagram of an electronic device suitable for implementing the six-degree-of-freedom treatment bed anti-collision method according to embodiments of the present disclosure is schematically shown.

[0112] As shown in Figure 3 The electronic device 300 according to embodiments of the present disclosure includes a processor 301 which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 302 or loaded from a storage portion 308 into a random access memory (RAM) 303. The processor 301 can include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset, and / or a special-purpose microprocessor (such as an application specific integrated circuit (ASIC)), etc. The processor 301 can also include an on-board memory for cache use. The processor 301 can include a single processing unit or multiple processing units for performing different actions of the method processes according to embodiments of the present disclosure.

[0113] In the RAM 303, various programs and data required for the operation of the electronic device 300 are stored. The processor 301, the ROM 302, and the RAM 303 are connected to each other via the bus 304. The processor 301 performs various operations of the method flow according to the embodiments of the present disclosure by executing the programs in the ROM 302 and / or the RAM 303. It should be noted that the programs can also be stored in one or more memories other than the ROM 302 and the RAM 303. The processor 301 can also perform various operations of the method flow according to the embodiments of the present disclosure by executing the programs stored in the one or more memories.

[0114] According to embodiments of the present disclosure, the electronic device 300 can further include an input / output (I / O) interface 305, which is also connected to the bus 304. The electronic device 300 can further include one or more of the following components connected to the input / output (I / O) interface 305: an input part 306 including a keyboard, a mouse, and the like; an output part 307 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage part 308 including a hard disk, and the like; and a communication part 309 including a network interface card such as a LAN card, a modem, and the like. The communication part 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the input / output (I / O) interface 305 as necessary. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 310 as necessary, so that a computer program read therefrom is installed in the storage part 308 as necessary.

[0115] The present disclosure also provides a computer readable storage medium, which can be included in the device / apparatus / system described in the above embodiments; or can exist separately without being assembled into the device / apparatus / system. The above computer readable storage medium carries one or more programs, when the one or more programs are executed, the method according to the embodiments of the present disclosure is implemented.

[0116] According to an embodiment of the present disclosure, the computer readable storage medium can be a nonvolatile computer readable storage medium, for example, can include, but is not limited to, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present disclosure, the computer readable storage medium can include the ROM 302 and / or the RAM 303 described above and / or one or more memory other than the ROM 302 and the RAM 303.

[0117] Embodiments of the present disclosure also include a computer program product, which includes a computer program containing program codes for executing the methods shown in the flowcharts. When the computer program product is run in a computer system, the program codes are used to make the computer system implement the six-degree-of-freedom treatment bed anti-collision method provided by the embodiments of the present disclosure.

[0118] The above-described functions defined in the system / apparatus / module / unit of the embodiments of the present disclosure are performed when the computer program is executed by the processor 301. According to an embodiment of the present disclosure, the system, apparatus, module, unit, etc. described above can be implemented by computer program modules.

[0119] In one embodiment, the computer program can rely on a tangible storage medium such as an optical storage device, a magnetic storage device, etc. In another embodiment, the computer program can also be transmitted, distributed, and downloaded in the form of a signal on a network medium and installed and downloaded through the communication part 309 and / or installed from the detachable medium 311. The program codes contained in the computer program can be transmitted by any appropriate network medium, including but not limited to wireless, wired, etc., or any appropriate combination thereof.

[0120] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 309 and / or installed from the detachable medium 311. When the computer program is executed by the processor 301, the above-described functions defined in the system of the embodiments of the present disclosure are performed. According to an embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by computer program modules.

[0121] According to embodiments of the present disclosure, program code of the computer programs provided by the embodiments of the present disclosure can be written in any combination of one or more programming languages, and specifically, these computer programs can be implemented using a high-level procedural and / or object-oriented programming language, and / or an assembly / machine language. The programming language includes, but is not limited to, a programming language such as Java, C++, Python, "C" language, or a similar programming language. The program code can be executed entirely on a user computing device, partially on a user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case involving a remote computing device, the remote computing device can be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, connected to the Internet through an Internet service provider).

[0122] The flow diagrams and the block diagrams in the drawings are illustrations of possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0123] Those skilled in the art can understand that the features described in various embodiments of the present disclosure can be combined and / or integrated in various combinations, even if such combinations are not explicitly described in the present disclosure. In particular, the features described in various embodiments of the present disclosure can be combined and / or integrated in various combinations without departing from the spirit and teachings of the present disclosure. All such combinations fall within the scope of the present disclosure.

[0124] The embodiments of the present disclosure are described above. However, these embodiments are merely for illustrative purposes, and are not intended to limit the scope of the present disclosure. Although each embodiment is described above separately, this does not mean that the measures in each embodiment cannot be used advantageously in combination. Without departing from the scope of the present disclosure, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications shall fall within the scope of the present disclosure.

Claims

1. A collision avoidance method for a six degree of freedom treatment couch based on point cloud, characterized in that, The method comprises the following steps: acquiring point cloud data of a static environment in a radiotherapy room, and generating a static point cloud model after point cloud reconstruction processing on the point cloud data; detecting the motion speed of a plurality of dynamic obstacles in the static environment, and classifying and collecting point cloud data of the dynamic obstacles according to a comparison result of the motion speed and a preset speed threshold; fusing point cloud data of a first dynamic obstacle in the plurality of dynamic obstacles with the static point cloud model to obtain a fused background point cloud model, the motion speed of the first dynamic obstacle being lower than the preset speed threshold; acquiring a local dynamic point cloud model of a six-degree-of-freedom treatment bed, calculating a minimum static distance between the local dynamic point cloud model and the background point cloud model, and a minimum dynamic distance between the local dynamic point cloud model and point cloud data of a second dynamic obstacle in the plurality of dynamic obstacles, the motion speed of the second dynamic obstacle being not lower than the preset speed threshold; controlling the motion state of the six-degree-of-freedom treatment bed according to a comparison result of the smaller value of the minimum static distance and the minimum dynamic distance and a safety distance threshold.

2. The method of claim 1, wherein, The detection of the motion speed of the plurality of dynamic obstacles in the static environment and the classification and collection of the point cloud data of the dynamic obstacles according to the comparison result of the motion speed and the preset speed threshold comprise: when the motion speed of any dynamic obstacle in the plurality of dynamic obstacles is lower than the preset motion threshold, the corresponding dynamic obstacle is determined as the first dynamic obstacle, and the point cloud data of the first dynamic obstacle is obtained through low-frequency scanning; when the motion speed of any dynamic obstacle in the plurality of dynamic obstacles is higher than or equal to the preset motion threshold, the corresponding dynamic obstacle is determined as the second dynamic obstacle, and the point cloud data of the second dynamic obstacle is obtained through a sensing device.

3. The method of claim 1, wherein, The fusion of the point cloud data of the first dynamic obstacle in the plurality of dynamic obstacles with the static point cloud model to obtain the background point cloud model comprises: registering the point cloud data of the first dynamic obstacle to the coordinate system of the static point cloud model through a rigid body transformation matrix; identifying an overlapping area of the point cloud data of the first dynamic obstacle and the static point cloud model, when the overlapping area is identified, reconstructing the static point cloud model according to the point cloud data in the overlapping area to obtain the background point cloud model; when the overlapping area cannot be identified, superimposing the point cloud data of the first dynamic obstacle into the static point cloud model to obtain the background point cloud model.

4. The method of claim 3, wherein, The reconstruction of the static point cloud model according to the point cloud data in the overlapping area to obtain the background point cloud model comprises: performing weighted fusion of the point cloud data in the overlapping area and the static point cloud model to obtain fused point cloud data; reconstructing the fused point cloud data to obtain the background point cloud model.

5. The method of claim 1, wherein, The six-degree-of-freedom treatment bed is composed of n treatment bed sub-components, wherein n = 1, 2, 3, …, N, and before the acquisition of the local dynamic point cloud model of the six-degree-of-freedom treatment bed, the method further comprises: The sub-point cloud data of n treatment bed sub-components in the local coordinate system is collected in real time by a motion sensor, the n treatment bed sub-components are connected according to a motion chain relationship, and the parent index of the nth treatment bed sub-component is n-1; For the sub-point cloud data of the nth treatment bed sub-component, the following operations are performed: The rotation amount and the translation amount of the nth treatment bed sub-component relative to the corresponding parent of the treatment bed sub-component are obtained by the motion sensor; The pose transformation matrix is determined according to the rotation amount and the translation amount; The motion transformation chain calculation is performed on the sub-point cloud data based on the coordinate transformation matrix and the pose transformation matrix, and the point cloud data in the coordinate system in the target space is obtained, wherein the coordinate transformation matrix is determined according to the position relationship of the six-degree-of-freedom treatment bed relative to the coordinate system in the radiotherapy room; When the index n of the treatment bed sub-component is incremented to N, the N corresponding point cloud data are taken as the point cloud data of the six-degree-of-freedom treatment bed.

6. The method of claim 1, wherein, The calculation of the minimum static distance between the local dynamic point cloud model and the background point cloud model, and the minimum dynamic distance between the local dynamic point cloud model and the point cloud data of the second dynamic obstacle in the plurality of dynamic obstacles includes: The spatial index structure of the background point cloud model and the spatial index structure of the second dynamic obstacle are constructed; The point cloud data in the local dynamic point cloud model are taken as a query point set, the query point set includes m point cloud data, m=1, 2, 3,…M, and the following operations are performed for the mth point cloud data in the query point set: The nearest neighbor point of the ith point cloud data in the background point cloud model is determined through the spatial index structure of the background point cloud model; The Euclidean distance between the mth point cloud data and the nearest neighbor point is calculated; When the index m of the query point set is incremented to M, the minimum value of the M Euclidean distances is taken as the minimum static distance; The nearest neighbor point of the mth point cloud data in the point cloud data of the second dynamic obstacle is determined through the spatial index structure of the second dynamic obstacle; The Euclidean distance between the mth point cloud data and the corresponding nearest neighbor point is calculated; When the index m of the query point set is incremented to M, the minimum value of the M Euclidean distances is taken as the minimum dynamic distance.

7. The method of claim 1, wherein, The comparison result of the smaller value of the minimum static distance and the minimum dynamic distance with the safety distance threshold value controls the motion state of the six-degree-of-freedom treatment bed, including: The smaller value of the minimum static distance and the minimum dynamic distance is taken as a predicted distance value; When the predicted distance is less than or equal to the safety distance threshold value, a first-level control instruction is generated to control the six-degree-of-freedom treatment bed to terminate motion; When the predicted distance is greater than the safety distance threshold value, a second-level control instruction is generated to control the six-degree-of-freedom treatment bed to decelerate motion or issue a warning signal.

8. A collision avoidance device for a six degree of freedom treatment couch based on point clouds, characterized by, It includes: A cloud model generation module obtains point cloud data of a static environment in a radiotherapy room, performs point cloud reconstruction processing on the point cloud data, and generates a static point cloud model. The classification acquisition module detects the movement speed of the plurality of dynamic obstacles in the static environment, and acquires point cloud data of the dynamic obstacles according to a comparison result of the movement speed and a preset speed threshold; The cloud model updating module fuses point cloud data of a first dynamic obstacle in the plurality of dynamic obstacles with the static point cloud model to obtain a fused background point cloud model, and the movement speed of the first dynamic obstacle is lower than the preset speed threshold; The distance calculation module obtains a local dynamic point cloud model of a six-degree-of-freedom treatment bed, calculates a minimum static distance between the local dynamic point cloud model and the background point cloud model, and a minimum dynamic distance between the local dynamic point cloud model and point cloud data of a second dynamic obstacle in the plurality of dynamic obstacles, and the movement speed of the second dynamic obstacle is not lower than the preset speed threshold; The operation control module controls a movement state of the six-degree-of-freedom treatment bed according to a comparison result of a smaller value of the minimum static distance and the minimum dynamic distance and a safety distance threshold. 9.An electronic device, comprising: one or more processors; a memory for storing one or more computer programs, characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, a computer program or instructions stored thereon, which are executed by a processor to implement the steps of the method according to any one of claims 1-7.

Images