A kind of automatic control method and system

Abstract

The present application discloses an automatic control method and system, the method comprising: identifying the three-dimensional features of the target object collected by the laser radar device; constructing the search index of the object according to the three-dimensional features of each object; searching whether there is a similar object similar to the object in the historical database according to the search index of the object; if there is, extracting the three-dimensional spatial feature attributes corresponding to the similar object from the historical database; creating a scaling model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all similar objects; collecting the movement information of the target object in real time through the radar device, and synchronizing the movement information in real time to the scaling model corresponding to the target object for simulation, and obtaining the simulation result of the target object; controlling the target object according to the simulation result. Thus, the problem of loss of details of the target object and difficulty in identifying the target object is effectively solved.

Description

Technical Field

[0001] The present application relates to the field of automation control technology, and in particular to an automation control method and system. Background Art

[0002] Nowadays, various industries are increasingly inclined to use LiDAR devices to replace traditional camera devices for the identification and control of target objects. This is mainly because LiDAR devices can not only effectively identify the three-dimensional geometric structure of target objects, but also perform basic operations such as distance measurement and speed measurement. These characteristics give LiDAR devices significant advantages in complex control tasks.

[0003] Due to the inherent characteristics of LiDAR, the radar point cloud generated by LiDAR is usually missing, resulting in the loss of certain detailed features of the target object. At the same time, when controlling the target object, it is generally difficult to use LiDAR to process small or delicate operations, which greatly affects the usability of LiDAR. Summary of the invention

[0004] Based on the above-mentioned deficiencies of the prior art, the present application provides an automatic control method and system to solve the problem of target object detail loss caused by the prior art.

[0005] In order to achieve the above objectives, this application provides the following technical solutions:

[0006] The first aspect of the present application provides an automatic control method, comprising:

[0007] Identify the three-dimensional features of a target object collected by a laser radar device; wherein the target object includes at least a plurality of objects;

[0008] For each of the objects, construct a search index for the object according to the three-dimensional features of the object;

[0009] searching, according to the search index of the object, whether there is a similar object similar to the object in the historical database;

[0010] If there is a similar object similar to the object in the historical database, extracting the three-dimensional spatial feature attributes corresponding to the similar object from the historical database;

[0011] According to the three-dimensional spatial feature attributes corresponding to all the similar objects, a scaled model corresponding to the target object is created using a multi-dimensional data space; wherein the distances between the respective proportional models included in the scaled model corresponding to the target object are obtained by proportionally converting the distances between the respective objects;

[0012] The movement information of the target object is collected in real time by a radar device, and the movement information is synchronized in real time to a scaling model corresponding to the target object for simulation, so as to obtain a simulation result of the target object;

[0013] The target object is controlled according to the simulation result.

[0014] Optionally, in the above-mentioned automatic control method, searching a history database for a similar object similar to the object according to the search index of the object includes:

[0015] Searching a historical database for a three-dimensional feature that is consistent with the search index;

[0016] Calculating a matching value between the three-dimensional feature and the search index;

[0017] Determine whether the matching value is not less than a preset threshold;

[0018] If the matching value is not less than the preset threshold, it is determined that there is a similar object similar to the object in the historical database.

[0019] Optionally, in the above-mentioned automatic control method, creating a scaling model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all the similar objects includes:

[0020] For all the similar objects, the three-dimensional spatial feature attributes corresponding to the similar objects are loaded into the multidimensional data space, and proportional scaling models corresponding to the similar objects are created based on the three-dimensional spatial feature attributes corresponding to the similar objects through the multidimensional data space;

[0021] assigning the spatial attributes of the object corresponding to the similar object to the proportional model;

[0022] Based on all the assigned proportional models, a scaled model corresponding to the target object is created.

[0023] Optionally, in the above-mentioned automatic control method, controlling the target object according to the simulation result includes:

[0024] The simulation result is fed back to the automatic control program, which analyzes the simulation result to obtain an execution instruction, and controls the target object to perform an operation corresponding to the execution instruction according to the execution instruction.

[0025] Optionally, in the above-mentioned automatic control method, it also includes:

[0026] In the process of controlling the target object, detecting in real time whether the target object is lost in tracking;

[0027] If the target object tracking is lost, replacing the radar device with a tracking device;

[0028] Using the tracking device to search for the target object, and detecting whether the target object is found;

[0029] If the target object is not found, obtaining the target movement information collected just before the target object is lost;

[0030] The target movement information is loaded into the scaling model contained in the multi-dimensional data space, so that the scaling model simulates the interaction between the various objects contained in the target object in real time according to the target movement information.

[0031] Optionally, in the above-mentioned automatic control method, it also includes:

[0032] Capture interactive behaviors in real time;

[0033] If there is a behavior in the interactive behavior that indicates that each of the objects is in a dangerous state during movement, a warning message is fed back to the front end, and the safe operation of the target object during movement at a future moment is simulated through the scaling model;

[0034] Feedback the simulated safety operations to the automated control program.

[0035] Optionally, in the above-mentioned automatic control method, after searching the history database for a similar object similar to the object according to the search index of the object, the method further comprises:

[0036] If there is no similar object similar to the object in the historical database, the three-dimensional features of the target object collected by the recognition laser radar device are returned.

[0037] A second aspect of the present application provides an automated control system, comprising:

[0038] An identification unit, used to identify three-dimensional features of a target object collected by a laser radar device; wherein the target object includes at least a plurality of objects;

[0039] A construction unit, configured to construct a search index for each of the objects according to the three-dimensional features of the object;

[0040] A searching unit, configured to search a history database for similar objects similar to the object according to a search index of the object;

[0041] an extraction unit, configured to extract a three-dimensional spatial feature attribute corresponding to a similar object from the historical database if there is a similar object similar to the object in the historical database;

[0042] A creation unit, used to create a scaled model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all the similar objects; wherein the distances between the respective proportional models included in the scaled model corresponding to the target object are obtained by proportionally converting the distances between the respective objects;

[0043] A simulation unit, used to collect the movement information of the target object in real time through a radar device, and synchronize the movement information in real time to a scaling model corresponding to the target object for simulation, so as to obtain a simulation result of the target object;

[0044] A control unit is used to control the target object according to the simulation result.

[0045] Optionally, in the above-mentioned automatic control system, the search unit includes:

[0046] A search subunit, used for searching a historical database for a three-dimensional feature consistent with the search index;

[0047] A calculation unit, used for calculating a matching value between the three-dimensional feature and the search index;

[0048] A judging unit, used to judge whether the matching value is not less than a preset threshold;

[0049] A determination unit is configured to determine that a similar object similar to the object exists in the history database if the matching value is not less than the preset threshold.

[0050] Optionally, in the above-mentioned automated control system, the creation unit includes:

[0051] A creation subunit is used to load the three-dimensional spatial feature attributes corresponding to all the similar objects into the multidimensional data space respectively, and create proportional scaling models corresponding to the similar objects based on the three-dimensional spatial feature attributes corresponding to the similar objects through the multidimensional data space;

[0052] an assignment unit, configured to assign the spatial attribute of the object corresponding to the similar object to the proportional model;

[0053] The model creation unit is used to create a scaled model corresponding to the target object based on all the assigned proportional models.

[0054] Optionally, in the above-mentioned automatic control system, the control unit includes:

[0055] The control subunit is used to feed back the simulation result to the automatic control program, and the automatic control program analyzes the simulation result to obtain an execution instruction, and controls the target object to execute the operation corresponding to the execution instruction according to the execution instruction.

[0056] Optionally, in the above-mentioned automatic control system, it also includes:

[0057] A first detection unit, used for detecting in real time whether the target object is lost in the process of controlling the target object;

[0058] a replacement unit, configured to replace the radar device with a tracking device if the tracking of the target object is lost;

[0059] A second detection unit, configured to use the tracking device to search for the target object and detect whether the target object is found;

[0060] an acquisition unit, configured to acquire target movement information collected just before the target object is lost if the target object is not found;

[0061] The interaction simulation unit is used to load the target movement information into the scaling model contained in the multi-dimensional data space, so that the scaling model simulates the interaction between each object contained in the target object in real time according to the target movement information.

[0062] Optionally, in the above-mentioned automatic control system, it also includes:

[0063] A behavior acquisition unit, used to acquire interactive behaviors in real time;

[0064] an operation simulation unit, configured to feed back warning information to the front end if there is a behavior in the interactive behavior indicating that each of the objects is in a dangerous state during movement, and to simulate the safe operation of the target object during movement at a future moment through the scaling model;

[0065] The feedback unit is used to feed back the simulated safety operation to the automation control program.

[0066] Optionally, in the above-mentioned automatic control system, it also includes:

[0067] An execution unit is used to return to execute the three-dimensional features of the target object collected by the recognition laser radar device if there is no similar object similar to the object in the historical database.

[0068] An automated control method provided by the present application identifies the three-dimensional features of a target object collected by a laser radar device, wherein the target object includes at least a plurality of objects, and then constructs a search index for each object according to the three-dimensional features of the object, and then searches a historical database for similar objects according to the search index of the object, and if there are similar objects similar to the object in the historical database, extracts the three-dimensional spatial feature attributes corresponding to the similar objects from the historical database, and then creates a scaled model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all similar objects, wherein the distances between the proportional models contained in the scaled model corresponding to the target object are obtained by proportionally converting the distances between the objects, and then collects the movement information of the target object in real time through the radar device, and synchronizes the movement information in real time to the scaled model corresponding to the target object for simulation, to obtain a simulation result of the target object, and finally controls the target object according to the simulation result. Therefore, the corresponding three-dimensional spatial feature attributes are obtained from the historical database through the three-dimensional point cloud data of the target object, and a scaling model of the target object is constructed. At the same time, the movement trajectory of the target object is estimated through the scaling model, and then the target object is automatically controlled according to the estimated results. Therefore, the problem of loss of target object details and the problem of difficulty in identifying the target object are effectively solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0070] Figure 1 A schematic diagram of a flow chart of an automated control method provided in an embodiment of the present application;

[0071] Figure 2 A schematic diagram of a flow chart of a method for searching similar objects provided in an embodiment of the present application;

[0072] Figure 3 A schematic diagram of a process for creating a scaling model provided in an embodiment of the present application;

[0073] Figure 4 A schematic flow chart of a method for tracking a target object is provided for another embodiment of the present application;

[0074] Figure 5 A schematic diagram of the structure of an automated control system provided in another embodiment of the present application. DETAILED DESCRIPTION

[0075] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0076] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0077] The present application embodiment provides an automated control method, such as Figure 1 As shown, the specific steps include:

[0078] S101. Identify the three-dimensional features of the target object collected by the laser radar device.

[0079] The target object includes at least a plurality of objects.

[0080] It is understandable that the target object may refer to a means of transportation such as a train or a car. When the target object refers to a train, the multiple objects contained in the target object refer to the carriages of the train. When the target object refers to a car, the multiple objects contained in the target object refer to other cars within a certain range from the car.

[0081] It should be noted that in order to achieve automated control of the target object, complete specified actions and interactive actions, and effectively avoid the occurrence of certain dangerous behaviors, the embodiment of the present application arranges a laser radar at an effective location and then sets the capture range of the laser radar equipment. Secondly, when the target object appears in the capture range, the laser radar will automatically capture three-dimensional point cloud data, and finally the three-dimensional features of the captured target object are identified through the three-dimensional point cloud data.

[0082] In addition, when the target object is a train, the valid location is directly above or to the side of the track. When the target object is a car, the valid location is directly above or to the side of the road, so the valid location is determined according to the type of the target object.

[0083] S102 : For each object, construct a search index for the object according to the three-dimensional features of the object.

[0084] It should be noted that since the three-dimensional features of the target object identified through three-dimensional point cloud data may not be complete enough, in order to read the complete three-dimensional wheel hub of the target object from the historical database for subsequent simulation of the target object's behavior, it is necessary to establish a search index for the target object in advance, that is, a search index for each object contained in the target object.

[0085] Specifically, a search index for the object may be constructed based on geometric features of the object, such as length, width, and height, or three-dimensional model features of the object.

[0086] S103 . Search the history database for similar objects similar to the object according to the search index of the object.

[0087] It is understandable that, since the historical database contains complete three-dimensional wheels of many different types of objects, it is possible to flexibly and conveniently search for objects similar to the object from the historical database based on the pre-built search index, and then obtain the complete three-dimensional profile if it can be found, thereby saving time and cost, and also improving accuracy. Therefore, if a similar object similar to the object is found in the historical database based on the search index of the object, it means that the historical database has a complete three-dimensional profile of the object, so step S104 is executed.

[0088] Optionally, after executing step S103, the method further includes:

[0089] If there is no similar object similar to the object in the historical database, the three-dimensional features of the target object collected by the laser radar device are returned for execution recognition.

[0090] It is understandable that when there is no similar object in the historical database, it means that the historical database does not have the complete three-dimensional outline of the object, so it is necessary to return to step S101 and continue to search for the three-dimensional point cloud data of the target object through the lidar.

[0091] Optionally, in another embodiment of the present application, a specific implementation of step S103 is as follows: Figure 2 As shown, the following steps are included:

[0092] S201. Search a historical database for three-dimensional features that are consistent with the search index.

[0093] Optionally, the search index may be compared with each three-dimensional feature in the historical database, or the same or similar feature extraction method may be used to generate a feature descriptor, and then the historical database may be searched for three-dimensional features consistent with the description of the feature descriptor.

[0094] S202: Calculate the matching value between the three-dimensional feature and the search index.

[0095] It can be understood that in order to know whether the found three-dimensional features match the search index, that is, to know whether the found three-dimensional features match the object corresponding to the search index, SIFT, SURF, and ORB matching algorithms can be used to compare the three-dimensional data with the search index, that is, to obtain the matching value by comparing the similarity of feature vectors or calculating the distance measurement between features.

[0096] S203: Determine whether the matching value is not less than a preset threshold.

[0097] Specifically, the system can set a matching threshold to obtain the degree of matching between the found three-dimensional feature and the object corresponding to the search index, that is, by judging whether the matching value is not less than the preset threshold. Therefore, if the matching value is not less than the preset threshold, it means that the found three-dimensional feature can match the object corresponding to the search index, so step S204 is executed.

[0098] Optionally, if the matching value is not less than a preset threshold, it means that the found three-dimensional feature cannot match the object corresponding to the search index, and therefore the process returns to step S101.

[0099] Optionally, the preset threshold is 8, and of course it can also be other thresholds, which are set according to specific needs.

[0100] S204: Determine whether there is a similar object similar to the object in the history database.

[0101] Specifically, when the matching value is not less than a preset threshold, it means that the found three-dimensional feature can be basically determined to be the three-dimensional feature of the object corresponding to the search index. At this time, it can be determined that there are similar objects similar to the object in the historical database.

[0102] S104: extracting three-dimensional spatial feature attributes corresponding to similar objects from the historical database.

[0103] Optionally, the three-dimensional spatial feature attributes corresponding to the extracted similar objects may be preprocessed, wherein the preprocessing includes feature normalization, dimensionality reduction processing (such as principal component analysis), data visualization, etc., so as to better understand the three-dimensional spatial feature attributes.

[0104] S105 . Create a scaling model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all similar objects.

[0105] It can be understood that the three-dimensional spatial feature attributes corresponding to all similar objects are substituted into the multidimensional data space, so as to simulate the scaled model of the target object in proportion according to the three-dimensional spatial features corresponding to all similar objects through the multidimensional data space, and to simulate the movement of the target object through the scaled model to avoid collision and other problems. Among them, the multidimensional data space is a spatial coordinate system with set xyz coordinate axes, and each coordinate scale represents a spatial node.

[0106] It should be noted that the distances between the various proportional models contained in the scaled model corresponding to the target object are obtained by proportionally converting the distances between the various objects, so that the scaled model corresponding to the target object fits the actual target object with high precision, thereby achieving the purpose of efficient and automated control of the target object.

[0107] Optionally, in another embodiment of the present application, a specific implementation of step S105 is as follows: Figure 3 As shown, the following steps are included:

[0108] S301, for all similar objects, loading the three-dimensional spatial feature attributes corresponding to the similar objects into the multidimensional data space, and creating proportional scaling models corresponding to the similar objects based on the three-dimensional spatial feature attributes corresponding to the similar objects through the multidimensional data space.

[0109] Specifically, a blank task can be created in a multidimensional data space first, and then the three-dimensional spatial feature attributes corresponding to similar objects can be loaded into the blank task in turn. Through the depiction function of the multidimensional data space, three-dimensional shapes can be depicted one by one according to the three-dimensional spatial feature attributes corresponding to each similar object. At this time, the drawn three-dimensional shape is the proportional scale model corresponding to the created similar object.

[0110] For example, in a multidimensional data space, the coordinate values ​​of the coordinate areas where similar objects exist are set to 1, while the coordinate values ​​of the blank areas are set to 0, so that the geometric outlines of similar objects can be completely drawn using 0 and 1 in the empty multidimensional data space.

[0111] S302, assigning spatial attributes of objects corresponding to similar objects to the proportional model.

[0112] Specifically, in order to make the proportional scaling model corresponding to the created similar object able to completely restore the data of the target object, it is also necessary to assign the spatial attributes of each object to the corresponding proportional model for fine-tuning, so as to obtain a highly restored proportional model of each object.

[0113] Among them, spatial attributes may include target three-dimensional structure, distance, speed, moving angle, etc.

[0114] S303: Create a scaled model corresponding to the target object based on all the assigned proportional models.

[0115] Optionally, each assigned proportional model may be placed in the multi-dimensional data space in sequence according to the position and distance of each object in the target object in the real scene, so that a scaled model corresponding to the target object can be obtained.

[0116] S106, collecting movement information of the target object in real time through the radar device, and synchronizing the movement information in real time to the scaling model corresponding to the target object for simulation, so as to obtain a simulation result of the target object.

[0117] It should be noted that in order to adjust the movement of the target object through the simulation results obtained by simulating the scaling model in the multidimensional data space to avoid collision and other problems, it is necessary to track the target object in real time through the radar equipment, collect the movement information of the target object in real time, and load the movement information into the scaling model corresponding to the target object in real time, so as to perform overall displacement of the values ​​on the coordinate points in the scaling model according to the movement information, such as moving from (x0, y0, z0) to (x1, y1, z1), that is, perform matrix operations on the spatial area of ​​the target object, and then estimate the displacement direction, movement trajectory and other information of the target object. When the trajectories overlap, the collision time of the target can be estimated. The loading rate of objects such as coal in the truck compartment can also be estimated according to the rising rate of the plane through the scaling model, and it can be estimated how long it will take for spillage to occur. The simulation results can be the displacement direction, movement trajectory, collision time, and how long it will take for spillage to occur, etc.

[0118] Optionally, the radar equipment may include laser radar, 4D millimeter wave radar, infrared radar and other equipment.

[0119] S107: Control the target object according to the simulation result.

[0120] It is understandable that when the simulation result indicates that the target object will not collide, the target object is controlled to travel. When the simulation result indicates that the target object may collide, the target object is controlled to stop traveling.

[0121] Optionally, in another embodiment of the present application, a specific implementation of step S107 includes the following steps:

[0122] The simulation results are fed back to the automation control program, which analyzes the simulation results, obtains execution instructions, and controls the target object to perform the operation corresponding to the execution instructions according to the execution instructions.

[0123] Specifically, the simulation results are fed back to the automation control program, which further analyzes the simulation results to determine whether the target object should continue the instruction or terminate the instruction or change the instruction.

[0124] For example, when the automation control program analyzes based on the simulation results that the target object will collide in a few minutes, the automation control program will generate a termination instruction, and then the automation control program controls the target object to terminate immediately according to the termination instruction.

[0125] Optionally, in the process of controlling the target object, the target object may be lost in tracking. In order to find the lost target object in time and avoid the occurrence of dangerous situations, in another embodiment of the present application, corresponding processing is also performed for the situation where the target object is lost in tracking. Figure 4 As shown, another embodiment of the present application provides a method for tracking a target object, comprising the following steps:

[0126] S401. During the process of controlling the target object, detecting in real time whether the target object is lost in tracking.

[0127] It is understandable that in the process of controlling the target object, there may be some interference factors that cause the target object to be lost. Therefore, in order to promptly know whether the target object is lost, it is possible to confirm by real-time detection of whether the radar device has collected the movement information of the target object. Therefore, if it is detected that the radar device has not collected the movement information of the target object, it means that the target object is lost, so execute step S402.

[0128] Optionally, if it is detected that the radar device has collected movement information of the target object, it means that the target object has not been lost in tracking, so the movement information is synchronized in real time to the scaling model corresponding to the target object for simulation to obtain the simulation result of the target object.

[0129] S402: Replace the radar device with a tracking device.

[0130] It is understandable that when the tracking of the target object is lost, the target object may have traveled out of the capture range of the radar device. Therefore, the radar device can be replaced with a tracking device with a longer capture range to track the target object.

[0131] S403: Use the tracking device to search for the target object and detect whether the target object is found.

[0132] Specifically, when the tracking device is replaced, it is necessary to immediately start the tracking device to scan whether there is a target object within the capture range, so as to obtain the moving trajectory of the target object in time.

[0133] Therefore, if the target object does not exist within the scanning capture range of the tracking device, it means that the target object cannot be found by the tracking device, so step S404 is executed at this time. Optionally, if the target object exists within the scanning capture range of the tracking device, it means that the target object can be found by the tracking device, so it is necessary to collect the movement information of the target object in real time through the tracking device, and synchronize the movement information in real time to the scaling model corresponding to the target object for simulation, so as to obtain the simulation result of the target object.

[0134] S404: Acquire target movement information collected just before the target object is lost.

[0135] It is understandable that when the target object is not found, it is necessary to obtain the target movement information of the target object last collected by the radar device, that is, the target movement information collected at the moment before the target object is lost, so as to subsequently simulate the displacement process of the target object at the future moment. Among them, the target movement information may include spatial position, speed, movement direction and other information.

[0136] S405 , loading the target movement information into the scaling model included in the multi-dimensional data space, so that the scaling model simulates the interaction between various objects included in the target object in real time according to the target movement information.

[0137] Specifically, for the safety of the target object, it is necessary to simulate the possible situations that may occur to the target object in the future, that is, the acquired target movement information needs to be loaded into the scaling model contained in the multidimensional data space, and the spatial displacement simulation is performed according to the target movement information through the scaling model, so as to simulate the interaction behavior between the various objects contained in the target object, so as to know whether the target object has dangerous behavior from the interactive behavior.

[0138] For example, in production sites such as autonomous driving and automatic loading, the displacement state of the target object is unpredictable. Therefore, we can only calculate whether the target object has collisions and other problems at certain time points based on the target movement information of the target object. Or when the production site is controlled by a robotic arm, because the robotic arm is fully controllable, the trajectory of the controlled target object is often fixed or completely stationary. At this time, the movement trajectory of the entire instruction can be completely simulated based on the control instructions of the robotic arm.

[0139] Optionally, based on Figure 4The interactive behavior simulated by the provided tracking method of a target object can determine whether the target object has dangerous behavior and respond in time according to the interactive behavior. Therefore, optionally, a safety prevention method for a target object provided in an embodiment of the present application includes the following steps:

[0140] Capture interactive behaviors in real time.

[0141] It is understandable that, since the target object has been lost from tracking, in order to grasp the situation of the target object in real time, it is necessary to obtain the interactive behavior simulated in step S405 in real time, so as to learn from the interactive behavior whether the target object is currently in a dangerous situation.

[0142] It should be noted that after obtaining the interaction behaviors between the various objects contained in the target object, it is necessary to determine the safety conditions of the target object based on the interaction behaviors, that is, to detect whether there are behaviors in the interaction behaviors that indicate that the various objects are in a dangerous state during movement. If there are behaviors in the interaction behaviors that indicate that the various objects are in a dangerous state during movement, early warning information is fed back to the front end, and the safe operation of the target object during movement at a future moment is simulated through the scaling model.

[0143] It should also be noted that when there are behaviors in the interactive behavior that indicate that each object is in a dangerous state during movement, the warning information needs to be fed back to the front end so that the front end can respond in time to avoid causing a larger accident. In addition, the scaling model can be used to simulate safe operations in dangerous states based on the currently known target movement information and interactive behaviors, so that the behavior of the target object in a dangerous state can be resolved based on the subsequent safe operations.

[0144] Feedback the simulated safety operations to the automated control program.

[0145] It is understandable that after obtaining safe operation, it is necessary to feed back the safe operation to the automation control program so as to control the target object to perform safe operation through the automation control program and prevent the occurrence of dangerous conditions.

[0146] For example, when there is a risk of vehicle collision or material spillage in the interactive behavior, the safety operation can be forced braking or valve closing, thereby controlling the target object to perform operations such as forced braking or valve closing through an automated control program.

[0147] An automated control method provided by the present application identifies the three-dimensional features of a target object collected by a laser radar device, wherein the target object includes at least a plurality of objects, and then constructs a search index for each object according to the three-dimensional features of the object, and then searches a historical database for similar objects according to the search index of the object, and if there are similar objects similar to the object in the historical database, extracts the three-dimensional spatial feature attributes corresponding to the similar objects from the historical database, and then creates a scaled model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all similar objects, wherein the distances between the proportional models contained in the scaled model corresponding to the target object are obtained by proportionally converting the distances between the objects, and then collects the movement information of the target object in real time through the radar device, and synchronizes the movement information in real time to the scaled model corresponding to the target object for simulation, to obtain a simulation result of the target object, and finally controls the target object according to the simulation result. Therefore, the corresponding three-dimensional spatial feature attributes are obtained from the historical database through the three-dimensional point cloud data of the target object, and a scaling model of the target object is constructed. At the same time, the movement trajectory of the target object is estimated through the scaling model, and then the target object is automatically controlled according to the estimated results. Therefore, the problem of loss of target object details and the problem of difficulty in identifying the target object are effectively solved.

[0148] Another embodiment of the present application provides an automated control system, such as Figure 5 As shown, it includes the following units:

[0149] The identification unit 501 is used to identify the three-dimensional features of the target object collected by the laser radar device. The target object includes at least a plurality of objects.

[0150] The construction unit 502 is used to construct a search index of each object according to the three-dimensional features of the object.

[0151] The search unit 503 is used to search the history database for similar objects similar to the object according to the search index of the object.

[0152] The extraction unit 504 is configured to extract the three-dimensional spatial feature attributes corresponding to the similar object from the historical database if there is a similar object similar to the object in the historical database.

[0153] The creation unit 505 is used to create a scaling model corresponding to the target object using the multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all similar objects, wherein the distances between the proportional models included in the scaling model corresponding to the target object are obtained by proportionally converting the distances between the various objects.

[0154] The simulation unit 506 is used to collect the movement information of the target object in real time through the radar equipment, and synchronize the movement information in real time to the scaling model corresponding to the target object for simulation to obtain the simulation result of the target object.

[0155] The control unit 507 is used to control the target object according to the simulation result.

[0156] It should be noted that the specific working process of the above-mentioned units in the embodiment of the present application can refer to steps S101 to S107 in the above-mentioned method embodiment, and will not be repeated here.

[0157] Optionally, in an automation control system provided in another embodiment of the present application, the search unit 503 includes:

[0158] The search subunit is used to search the 3D features consistent with the search index from the history database.

[0159] The calculation unit is used to calculate the matching value between the three-dimensional feature and the search index.

[0160] The judging unit is used to judge whether the matching value is not less than a preset threshold.

[0161] The determination unit is configured to determine that there is a similar object similar to the object in the history database if the matching value is not less than a preset threshold.

[0162] Optionally, in an automation control system provided in another embodiment of the present application, the creation unit 505 includes:

[0163] A sub-unit is created to load the three-dimensional spatial feature attributes corresponding to all similar objects into the multidimensional data space respectively, and to create proportional scaling models corresponding to the similar objects based on the three-dimensional spatial feature attributes corresponding to the similar objects through the multidimensional data space.

[0164] The assignment unit is used to assign the spatial properties of objects corresponding to similar objects to the proportional model.

[0165] The model creation unit is used to create a scaled model corresponding to the target object based on all the assigned proportional models.

[0166] Optionally, in an automation control system provided in another embodiment of the present application, the control unit 507 includes:

[0167] The control subunit is used to feed back the simulation results to the automatic control program, which analyzes the simulation results, obtains execution instructions, and controls the target object to perform operations corresponding to the execution instructions according to the execution instructions.

[0168] Optionally, another embodiment of the present application provides an automated control system, further comprising:

[0169] The first detection unit is used to detect in real time whether the target object is lost in the process of controlling the target object.

[0170] A replacement unit is used to replace the radar device with the tracking device if the tracking of the target object is lost.

[0171] The second detection unit is used to use the tracking device to find the target object and detect whether the target object is found.

[0172] The acquisition unit is used to acquire the target movement information collected just before the target object is lost if the target object is not found.

[0173] The interaction simulation unit is used to load the target movement information into the scaling model contained in the multi-dimensional data space, so that the scaling model simulates the interaction between the various objects contained in the target object in real time according to the target movement information.

[0174] Optionally, another embodiment of the present application provides an automated control system, further comprising:

[0175] The behavior acquisition unit is used to acquire interactive behaviors in real time.

[0176] The operation simulation unit is used to feed back warning information to the front end if there is any behavior in the interactive behavior that indicates that each object is in a dangerous state during movement, and simulate the safe operation of the target object during movement at a future moment through a scaling model.

[0177] The feedback unit is used to feed back the simulated safety operation to the automation control program.

[0178] Optionally, another embodiment of the present application provides an automated control system, further comprising:

[0179] The execution unit is used to return to execute the recognition of the three-dimensional features of the target object collected by the laser radar device if there is no similar object similar to the object in the historical database.

[0180] It should be noted that the specific working process of each unit provided in the above embodiments of the present application can refer to the corresponding steps in the above method embodiments, and will not be repeated here.

[0181] It should also be noted that an automated control system provided in an embodiment of the present application has the technical effects of any of the above embodiments, and the embodiments of the present application are not described in detail here.

[0182] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0183] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An automated control method, characterized in that: include: Identify the three-dimensional features of a target object collected by a laser radar device; wherein the target object includes at least a plurality of objects; For each of the objects, construct a search index for the object according to the three-dimensional features of the object; searching, according to the search index of the object, whether there is a similar object similar to the object in the historical database; If there is a similar object similar to the object in the historical database, extracting the three-dimensional spatial feature attributes corresponding to the similar object from the historical database; According to the three-dimensional spatial characteristic attributes corresponding to all the similar objects, a scaling model corresponding to the target object is created using a multidimensional data space, which includes: for all the similar objects, the three-dimensional spatial characteristic attributes corresponding to the similar objects are loaded into the multidimensional data space respectively, and proportional scaling models corresponding to the similar objects are created through the multidimensional data space based on the three-dimensional spatial characteristic attributes corresponding to the similar objects; the spatial attributes of the objects corresponding to the similar objects are assigned to the proportional models; based on all the assigned proportional models, a scaling model corresponding to the target object is created; wherein the distance between each proportional model included in the scaling model corresponding to the target object is obtained by proportionally converting the distance between each of the objects; The movement information of the target object is collected in real time by a radar device, and the movement information is synchronized in real time to a scaling model corresponding to the target object for simulation, so as to obtain a simulation result of the target object; The target object is controlled according to the simulation result.

2. The method according to claim 1, characterized in that The searching, based on the search index of the object, whether there is a similar object similar to the object in the history database comprises: Searching a historical database for a three-dimensional feature that is consistent with the search index; Calculating a matching value between the three-dimensional feature and the search index; Determine whether the matching value is not less than a preset threshold; If the matching value is not less than the preset threshold, it is determined that there is a similar object similar to the object in the historical database.

3. The method according to claim 1, characterized in that The controlling the target object according to the simulation result includes: The simulation result is fed back to the automatic control program, which analyzes the simulation result to obtain an execution instruction, and controls the target object to perform an operation corresponding to the execution instruction according to the execution instruction.

4. The method according to claim 1, characterized in that Also includes: In the process of controlling the target object, detecting in real time whether the target object is lost in tracking; If the target object tracking is lost, replacing the radar device with a tracking device; Using the tracking device to search for the target object, and detecting whether the target object is found; If the target object is not found, obtaining the target movement information collected just before the target object is lost; The target movement information is loaded into the scaling model contained in the multi-dimensional data space, so that the scaling model simulates the interaction between the various objects contained in the target object in real time according to the target movement information.

5. The method according to claim 4, characterized in that Also includes: Capture interactive behaviors in real time; If there is a behavior in the interactive behavior that indicates that each of the objects is in a dangerous state during movement, a warning message is fed back to the front end, and the safe operation of the target object during movement at a future moment is simulated through the scaling model; Feedback the simulated safety operations to the automated control program.

6. The method according to claim 1, characterized in that After searching the history database for similar objects similar to the object according to the search index of the object, the method further includes: If there is no similar object similar to the object in the historical database, the three-dimensional features of the target object collected by the recognition laser radar device are returned.

7. An automated control system, characterized in that: include: An identification unit, used to identify the three-dimensional features of a target object collected by a laser radar device; wherein the target object includes at least a plurality of objects; A construction unit, configured to construct a search index for each of the objects according to the three-dimensional features of the object; A searching unit, configured to search a history database for similar objects similar to the object according to a search index of the object; an extraction unit, configured to extract a three-dimensional spatial feature attribute corresponding to a similar object from the historical database if there is a similar object similar to the object in the historical database; A creation unit is used to create a scaling model corresponding to the target object using a multidimensional data space according to the three-dimensional spatial feature attributes corresponding to all the similar objects, which includes: for all the similar objects, the three-dimensional spatial feature attributes corresponding to the similar objects are loaded into the multidimensional data space, and the proportional scaling models corresponding to the similar objects are created through the multidimensional data space based on the three-dimensional spatial feature attributes corresponding to the similar objects; the spatial attributes of the objects corresponding to the similar objects are assigned to the proportional models; based on all the assigned proportional models, the scaling model corresponding to the target object is created; wherein the distances between the proportional models included in the scaling model corresponding to the target object are obtained by proportionally converting the distances between the objects; A simulation unit, used to collect the movement information of the target object in real time through a radar device, and synchronize the movement information in real time to a scaling model corresponding to the target object for simulation, so as to obtain a simulation result of the target object; A control unit is used to control the target object according to the simulation result.

8. The system according to claim 7, characterized in that The searching unit comprises: A search subunit, used for searching a historical database for a three-dimensional feature consistent with the search index; A calculation unit, used for calculating a matching value between the three-dimensional feature and the search index; A judging unit, used to judge whether the matching value is not less than a preset threshold; A determination unit is configured to determine that a similar object similar to the object exists in the history database if the matching value is not less than the preset threshold.

9. The system according to claim 7, characterized in that The creation unit comprises: A creation subunit is used to load the three-dimensional spatial feature attributes corresponding to all the similar objects into the multidimensional data space respectively, and create proportional scaling models corresponding to the similar objects based on the three-dimensional spatial feature attributes corresponding to the similar objects through the multidimensional data space; an assignment unit, configured to assign the spatial attribute of the object corresponding to the similar object to the proportional model; The model creation unit is used to create a scaled model corresponding to the target object based on all the assigned proportional models.

Images