Safe loading and unloading method and system based on laser radar

Abstract

The present invention relates to the field of data processing technology, and in particular to a method and system for safe loading and unloading of vehicles based on laser radar, comprising: obtaining point cloud data of the surrounding environment of the loading and unloading vehicle; obtaining the motion state factor of each obstacle at each acquisition moment; obtaining the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition moment; correcting the motion state factor of the obstacle according to the interaction factor between the obstacle and the loading and unloading vehicle to obtain the corrected motion state factor of each obstacle; obtaining the danger coefficient of each obstacle to the loading and unloading vehicle according to the corrected motion state factor; and performing safety detection on the loading and unloading vehicle according to the danger coefficient of the obstacle to the loading and unloading vehicle. The present invention improves the safety of the loading and unloading vehicle during loading and unloading.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a laser radar-based safe vehicle loading and unloading method and system. Background Art

[0002] The general safe loading and unloading method based on LiDAR is to obtain point cloud data of obstacles around the loading and unloading vehicle through LiDAR ranging, obtain the three-dimensional position data of the obstacles through analysis and processing of the point cloud data, and transmit the three-dimensional position data of the obstacles to the remote monitoring system, thereby realizing all-round monitoring and management of the loading and unloading operation process.

[0003] In a complex loading and unloading environment, there are a large number of dynamic objects around the loading and unloading truck, such as pedestrians and vehicles. The movement of the loading and unloading truck will cause the position of obstacles to change, and the obstacles may be blocked in the loading and unloading scene, resulting in obstacles in some areas cannot be correctly detected by the lidar, making the three-dimensional position data of the obstacles obtained by the lidar inaccurate, thereby reducing the safety of the loading and unloading truck during loading and unloading. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a safe vehicle loading and unloading method and system based on laser radar.

[0005] An embodiment of the present invention provides a method for safe loading and unloading of a vehicle based on a laser radar, the method comprising the following steps:

[0006] Acquire point cloud data of the surrounding environment of the loading and unloading vehicle; the point cloud data includes point cloud data sets of the loading and unloading vehicle at several acquisition moments and point cloud data sets of several obstacles;

[0007] According to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moment, the motion state factor of each obstacle at each acquisition moment is obtained; according to the relative position difference of the point cloud data set of each obstacle and the loading and unloading vehicle at each acquisition moment, the interaction factor of each obstacle and the loading and unloading vehicle at each acquisition moment is obtained;

[0008] The motion state factor of the obstacle is corrected according to the interaction factor between the obstacle and the loading and unloading vehicle, and the corrected motion state factor of each obstacle is obtained; according to the corrected motion state factor, the danger coefficient of each obstacle to the loading and unloading vehicle is obtained;

[0009] Safety checks are performed on loading and unloading vehicles based on the risk factors of obstacles to the loading and unloading vehicles.

[0010] Preferably, the method of obtaining the motion state factor of each obstacle at each acquisition moment according to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moment includes the following specific methods:

[0011] For any obstacle, the point cloud data set of the obstacle at each acquisition time is registered using the point cloud registration technology to obtain the movement direction of each point cloud data point in the point cloud data set of the obstacle at each acquisition time;

[0012] According to the difference in the movement direction of each point cloud data point at adjacent acquisition moments, the The first The cloud dataset of obstacles The motion trajectory factor of each point cloud data point;

[0013] According to the position difference of each obstacle at adjacent acquisition times, the The first The speed factor of the obstacle;

[0014] Get the The first The calculation method of the motion state factor of an obstacle is:

[0015]

[0016] In the formula, Indicates The first The motion state factor of each obstacle; Indicates The first The speed factor of the obstacle; Indicates The first The mean of the motion trajectory factors of all point cloud data points in the cloud dataset of obstacles.

[0017] Preferably, the obtaining The first The cloud dataset of obstacles The motion trajectory factor of the point cloud data point includes the following specific methods:

[0018] The first The first The cloud dataset of obstacles The unit vector in the vertical direction of the movement direction of the point cloud data point is used as the The first The normal unit vector of the point cloud data point; The first The normal unit vector of the point cloud data point is The first The normal unit vector of each point cloud data point is subtracted, and the modulus of the difference vector is obtained as the first The first The cloud dataset of obstacles The motion trajectory factor of each point cloud data point.

[0019] Preferably, the obtaining The first The speed factor of each obstacle includes the following specific methods:

[0020] The first The first The point cloud data point at the centroid position in the cloud data set of the obstacle is The first The Euclidean distance between the point cloud data points at the centroid position in the cloud data set of the obstacle is taken as the The first The speed factor of the obstacle.

[0021] Preferably, the method of obtaining the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition moment according to the relative position difference between each obstacle and the loading and unloading vehicle at each acquisition moment includes the following specific methods:

[0022] According to the position difference between each obstacle and the point cloud dataset of the loading and unloading vehicle at each acquisition time, the first The first The distance position difference factor of each obstacle;

[0023] According to the movement direction of the point cloud data point, get the The first The cloud dataset of obstacles The motion direction difference factor of each point cloud data point;

[0024] Get the The first The calculation method of the interaction factor between obstacles and loading and unloading vehicles is:

[0025]

[0026] In the formula, Indicates The first The interaction factor between obstacles and loading and unloading vehicles; Indicates The first The distance position difference factor of each obstacle; Indicates The first The mean of the motion direction difference factors of all point cloud data points in the cloud data set of obstacles; Represents an exponential function with a natural constant as base.

[0027] Preferably, the obtaining The first The distance position difference factor of each obstacle includes the following specific methods:

[0028] The first The first The point cloud data point at the centroid position in the cloud data set of the obstacle is The Euclidean distance between the point cloud data points at the centroid position in the cloud data set of the loading and unloading truck at the collection time is recorded as The first The first distance of the obstacle; The first The first distance to the obstacle and the The first The difference of the first distance of the obstacles is taken as the The first The distance position difference value of the obstacle; if the distance position difference value is greater than or equal to 0, 0 is used as the The first The distance position difference factor of the obstacle; if the distance position difference value is less than 0, the distance position difference value is used as the first The first The distance position difference factor of each obstacle.

[0029] Preferably, the obtaining The first The cloud dataset of obstacles The motion direction difference factor of each point cloud data point includes the following specific methods:

[0030] The first The first The cloud dataset of obstacles The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The first The cloud dataset of obstacles The movement direction value of each point cloud data point;

[0031] The first The cloud data set of loading and unloading vehicles at the collection time The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The cloud data set of loading and unloading vehicles at the collection time The movement direction value of each point cloud data point;

[0032] The first The first The cloud dataset of obstacles The motion direction value of the point cloud data point is The cloud data set of loading and unloading vehicles at the collection time The absolute value of the difference between the motion direction values ​​of the point cloud data points is taken as the The first The cloud dataset of obstacles The motion direction difference factor of each point cloud data point.

[0033] Preferably, the motion state factor of the obstacle is corrected according to the interaction factor between the obstacle and the loading and unloading vehicle to obtain the corrected motion state factor of each obstacle, including the specific method of:

[0034] The first The first The interaction factor between the obstacle and the loading and unloading vehicle is The first The product of the motion state factors of the obstacles is taken as the The first The corrected motion state factor of an obstacle.

[0035] Preferably, the method of obtaining the risk factor of each obstacle to the loading and unloading vehicle according to the corrected motion state factor includes:

[0036] The sequence of all the acquisition moments is recorded as the acquisition moment time series; The sequence of the corrected motion state factors of obstacles is recorded as The motion sequence of obstacles; The product of the standard deviation of the motion sequence of the obstacle and the standard deviation of the time series at the time of acquisition is recorded as the first product; The covariance between the motion sequence of each obstacle and the time series of the acquisition moment is recorded as the first covariance. The ratio of the first covariance to the first product is recorded as the first The danger factor of an obstacle to loading and unloading vehicles.

[0037] The present invention also proposes a laser radar-based safe loading and unloading system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of the laser radar-based safe loading and unloading method when executing the computer program.

[0038] The beneficial effects of the technical solution of the present invention are as follows: the present invention obtains point cloud data of the surrounding environment of a loading and unloading vehicle; obtains the motion state factor of each obstacle at each acquisition moment; obtains the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition moment; corrects the motion state factor of the obstacle according to the interaction factor between the obstacle and the loading and unloading vehicle to obtain the corrected motion state factor of each obstacle; obtains the danger factor of each obstacle to the loading and unloading vehicle according to the corrected motion state factor; performs safety detection on the loading and unloading vehicle according to the danger factor of the obstacle to the loading and unloading vehicle; and obtains the specific motion state position of the obstacle relative to the loading and unloading vehicle, thereby avoiding the formation of an occlusion blind spot due to mutual occlusion between obstacles, which leads to the erroneous judgment of the safety of the loading and unloading vehicle by the obstacle, thereby improving the safety of the loading and unloading vehicle during loading and unloading. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 The present invention is a flowchart of the steps of the method for safely loading and unloading vehicles based on laser radar.

[0041] Figure 2 This is a characteristic relationship flow chart of the safe loading and unloading of vehicles based on laser radar in the present invention. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of the laser radar-based safe loading and unloading method and system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of the safe loading and unloading method and system based on laser radar provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of a method for safe loading and unloading of vehicles based on laser radar according to an embodiment of the present invention, the method comprising the following steps:

[0046] Step S001: Acquire point cloud data of the surrounding environment of the loading and unloading vehicle.

[0047] It should be noted that in this embodiment, due to the presence of dynamic objects in the surrounding environment of the loading and unloading truck, and their positions and states are constantly changing, which has a potential impact on the safety of the loading and unloading truck. Therefore, when using laser radar to scan the surrounding environment, more scanning points need to be allocated to dynamic objects, thereby improving the safety loading and unloading vehicle system's perception of the dynamic environment and reducing potential risks.

[0048] Specifically, we first need to collect point cloud data of the environment around the loading and unloading vehicle. The specific process is as follows:

[0049] The laser radar sensor is used to scan and measure the surrounding environment of the loading and unloading truck every 1 minute to obtain the point cloud data set of all obstacles in the surrounding environment of the loading and unloading truck. The data collection takes 2 hours in total. The point cloud data composed of the point cloud data set of all obstacles at each collection moment is used as the point cloud data of the surrounding environment of the loading and unloading truck.

[0050] The point cloud data of the surrounding environment of the loading and unloading vehicle includes point cloud data sets of the loading and unloading vehicle at several acquisition moments and point cloud data sets of several obstacles, and the point cloud data sets include several identical point cloud data points.

[0051] So far, the point cloud data of the surrounding environment of the loading and unloading vehicle is obtained through the above method.

[0052] Step S002: According to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moments, the motion state factor of each obstacle at each acquisition moment is obtained; according to the relative position difference of the point cloud data set of each obstacle and the loading and unloading vehicle at each acquisition moment, the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition moment is obtained.

[0053] 1. Obtain the motion state factor of each obstacle at each acquisition moment.

[0054] It should be noted that in the complex surrounding environment of the loading and unloading truck, the difference between the normal vectors of each point cloud data point in the point cloud data set of the obstacle at adjacent acquisition moments can be used to measure its motion state; the greater the difference, the more obvious the movement of the obstacle, which may enter the operating area of ​​the loading and unloading truck and bring potential safety hazards; in addition, the speed of the obstacle is also a key factor in evaluating its motion state; by analyzing the speed of the obstacle, its potential impact on the loading and unloading truck can be more accurately judged, and the motion state factor of the obstacle can be quantified by combining the difference between the point cloud data sets of the obstacle at adjacent acquisition moments with the speed obtained by the displacement and time interval of the obstacle between different acquisition moments.

[0055] Specifically, for any obstacle, the point cloud registration technology is used to register the point cloud data set of the obstacle at each acquisition time, and the movement direction of each point cloud data point in the point cloud data set of the obstacle at each acquisition time is obtained.

[0056] Among them, the point cloud registration technology is an existing technology and will not be described in detail in this embodiment.

[0057] The first The first The cloud dataset of obstacles The unit vector in the vertical direction of the movement direction of the point cloud data point is used as the The first The normal unit vector of the point cloud data point; The first The normal unit vector of the point cloud data point is The first The normal unit vector of each point cloud data point is subtracted, and the modulus of the difference vector is obtained as the first The first The cloud dataset of obstacles The motion trajectory factor of each point cloud data point.

[0058] The first The first The point cloud data point at the centroid position in the cloud data set of the obstacle is The first The Euclidean distance between the point cloud data points at the centroid position in the cloud data set of the obstacle is taken as the The first The speed factor of the obstacle.

[0059] According to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moments, the motion state factor of each obstacle at each acquisition moment is obtained.

[0060] As an example, get the The first The calculation method of the motion state factor of an obstacle is:

[0061]

[0062] In the formula, Indicates The first The motion state factor of each obstacle; Indicates The first The speed factor of the obstacle; Indicates The first The mean of the motion trajectory factors of all point cloud data points in the cloud dataset of obstacles.

[0063] It should be noted that The larger the motion trajectory factor of the point cloud data point is, the larger the The normal vectors of all the point cloud data points in the point cloud data set at adjacent acquisition times are quite different, which means that the The more obvious the movement trajectory of each obstacle is; The larger the speed factor of the first obstacle, the The movement of obstacles is more obvious and faster, and they are more likely to enter the working area of ​​the loading and unloading vehicles, bringing potential safety hazards. Each obstacle requires more scanning points to allow for more detailed attention to the The impact of obstacles on loading and unloading vehicles.

[0064] At this point, the motion state factor of each obstacle at each acquisition moment is obtained.

[0065] 2. Obtain the interaction factor between each obstacle and the loading and unloading vehicle at each collection moment.

[0066] It should be noted that for the specific dynamic obstacles obtained by the laser radar scanning, the difference in distance between the obstacle and the loading and unloading vehicle at adjacent moments and whether the movement directions of the two are moving in the same direction can be determined, and the two can be combined to obtain the interaction factor between the obstacle and the loading and unloading vehicle in the moving state; when the obstacle and the loading and unloading vehicle have strong interaction, the displacement changes of the two in time are close to the same direction, so the difference in the angle between the movement direction and the horizontal direction of all point cloud data points of the two and the horizontal direction of the direction vector of the center of mass connection of the two can be calculated, and it can be quantified as a parameter for judging whether the movement directions of the obstacle and the loading and unloading vehicle are approaching each other, that is, the interaction factor between the obstacle and the loading and unloading vehicle.

[0067] Specifically, the The first The point cloud data point at the centroid position in the cloud data set of the obstacle is The Euclidean distance between the point cloud data points at the centroid position in the cloud data set of the loading and unloading truck at the collection time is recorded as The first The first distance of the obstacle; The first The first distance to the obstacle and the The first The difference of the first distance of the obstacles is taken as the The first The distance position difference value of the obstacle; if the distance position difference value is greater than or equal to 0, 0 is used as the The first The distance position difference factor of the obstacle; if the distance position difference value is less than 0, the distance position difference value is used as the first The first The distance position difference factor of each obstacle.

[0068] The first The first The cloud dataset of obstacles The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The first The cloud dataset of obstacles The movement direction value of the point cloud data point; The cloud data set of loading and unloading vehicles at the collection time The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The cloud data set of loading and unloading vehicles at the collection time The movement direction value of the point cloud data point; The first The cloud dataset of obstacles The motion direction value of the point cloud data point is The cloud data set of loading and unloading vehicles at the collection time The absolute value of the difference between the motion direction values ​​of the point cloud data points is taken as the The first The cloud dataset of obstacles The motion direction difference factor of each point cloud data point.

[0069] According to the relative position difference between each obstacle and the loading and unloading vehicle in the point cloud data set at each acquisition time, the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition time is obtained.

[0070] As an example, get the The first The calculation method of the interaction factor between obstacles and loading and unloading vehicles is:

[0071]

[0072] In the formula, Indicates The first The interaction factor between obstacles and loading and unloading vehicles; Indicates The first The distance position difference factor of each obstacle; Indicates The first The mean of the motion direction difference factors of all point cloud data points in the cloud data set of obstacles; represents an exponential function with a natural constant as the base, and the embodiment adopts The model shows an inverse proportional relationship. As the input of the model, implementers can choose the inverse proportional function according to actual conditions.

[0073] It should be noted that When the distance position difference of the obstacle is positive, it means that The first obstacle is approaching the loading and unloading vehicle. On the contrary, when the value is negative, it means that the The first obstacle is moving away from the loading and unloading vehicle, the smaller the interactivity; the larger the difference factor of the moving direction, the smaller the interactivity. The displacement change of the first obstacle over time does not approach the center of the line connecting the obstacle and the loading and unloading vehicle, that is, the obstacle does not approach the loading and unloading vehicle. The interaction between the obstacle and the loading and unloading vehicle is low.

[0074] At this point, the interaction factor between each obstacle and the loading and unloading vehicle at each collection moment is obtained through the above method.

[0075] Step S003: Correct the motion state factor of the obstacle according to the interaction factor between the obstacle and the loading and unloading vehicle to obtain the corrected motion state factor of each obstacle; obtain the risk factor of each obstacle to the loading and unloading vehicle according to the corrected motion state factor.

[0076] 1. Obtain the corrected motion state factor of each obstacle at each acquisition moment.

[0077] It should be noted that if the motion state factor of the obstacle is large and the moving target is a loading and unloading truck, that is, the interaction factor with the loading and unloading truck is also large, it means that the potential impact of the obstacle on the loading and unloading truck should be greater than other obstacles, and it is necessary to use a higher density of laser radar scanning points to scan it; therefore, the motion state factor of the obstacle is corrected by the interaction factor with the loading and unloading truck.

[0078] Specifically, as an example, get the The first The calculation method of the corrected motion state factor of each obstacle is:

[0079] The first The first The interaction factor between the obstacle and the loading and unloading vehicle is The first The product of the motion state factors of the obstacles is taken as the The first The corrected motion state factor of an obstacle.

[0080] At this point, the corrected motion state factor of each obstacle at each acquisition moment is obtained.

[0081] 2. Obtain the risk factor of each obstacle to the loading and unloading vehicle.

[0082] It should be noted that the degree of correlation between the corrected motion state factor of the obstacle at all acquisition moments and the time change is judged, that is, the degree of influence of the time change on the corrected motion state factor. If the corrected motion state factor is highly correlated with the time change, it means that the movement of the obstacle is closely related to the time change, which may have a greater impact on the safety of the loading and unloading vehicle; conversely, if the correlation between the two is low, the impact of the obstacle's movement on the safety of the loading and unloading vehicle may be small.

[0083] Specifically, the sequence of all collection moments is recorded as the collection moment time series; The sequence of the corrected motion state factors of obstacles is recorded as The motion sequence of obstacles; The product of the standard deviation of the motion sequence of the obstacle and the standard deviation of the time series at the time of acquisition is recorded as the first product; The covariance between the motion sequence of each obstacle and the time series of the acquisition moment is recorded as the first covariance. The ratio of the first covariance to the first product is recorded as the first The danger factor of an obstacle to loading and unloading vehicles.

[0084] So far, the risk factor of each obstacle to the loading and unloading vehicle is obtained through the above method.

[0085] Step S004: Perform safety inspection on the loading and unloading vehicle according to the risk factor of obstacles to the loading and unloading vehicle.

[0086] Preset a threshold parameter , wherein this embodiment is based on This example is described as an example, and this embodiment is not specifically limited. It depends on the specific implementation situation.

[0087] Specifically, for any obstacle, if the risk factor of the obstacle to the loading and unloading vehicle is greater than or equal to the threshold parameter , an alarm is issued to the driver.

[0088] See also Figure 2 , which shows a characteristic relationship flow chart for safe loading and unloading of vehicles based on lidar.

[0089] Through the above steps, the safe loading and unloading method based on lidar is completed.

[0090] The present invention also proposes a laser radar-based safe loading and unloading system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the laser radar-based safe loading and unloading method in steps S001 to S004 are implemented.

[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A safe vehicle loading and unloading method based on laser radar, characterized in that: The method comprises the following steps: Acquire point cloud data of the surrounding environment of the loading and unloading vehicle; the point cloud data includes point cloud data sets of the loading and unloading vehicle at several acquisition moments and point cloud data sets of several obstacles; According to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moment, the motion state factor of each obstacle at each acquisition moment is obtained; according to the relative position difference of the point cloud data set of each obstacle and the loading and unloading vehicle at each acquisition moment, the interaction factor of each obstacle and the loading and unloading vehicle at each acquisition moment is obtained; The motion state factor of the obstacle is corrected according to the interaction factor between the obstacle and the loading and unloading vehicle, and the corrected motion state factor of each obstacle is obtained; according to the corrected motion state factor, the danger coefficient of each obstacle to the loading and unloading vehicle is obtained; Conduct safety inspection on loading and unloading vehicles according to the risk factor of obstacles to the loading and unloading vehicles; The method of obtaining the motion state factor of each obstacle at each acquisition moment according to the position distribution difference of the point cloud data set of each obstacle between each acquisition moment and the adjacent acquisition moment includes: For any obstacle, the point cloud data set of the obstacle at each acquisition time is registered using the point cloud registration technology to obtain the movement direction of each point cloud data point in the point cloud data set of the obstacle at each acquisition time; According to the difference in the movement direction of each point cloud data point at adjacent acquisition moments, the The first The point cloud dataset of obstacles The motion trajectory factor of each point cloud data point; According to the position difference of each obstacle at adjacent acquisition times, the The first The speed factor of each obstacle; Get the The first The calculation method of the motion state factor of an obstacle is: In the formula, Indicates The first The motion state factor of each obstacle; Indicates The first The speed factor of the obstacle; Indicates The first The mean of the motion trajectory factors of all point cloud data points in the point cloud data set of obstacles; The acquisition The first The point cloud dataset of obstacles The motion trajectory factor of the point cloud data point includes the following specific methods: The first The first The point cloud dataset of obstacles The unit vector in the vertical direction of the movement direction of the point cloud data point is used as the The first The normal unit vector of the point cloud data point; The first The normal unit vector of the point cloud data point is The first The normal unit vector of each point cloud data point is subtracted, and the modulus of the difference vector is obtained as the first The first The point cloud dataset of obstacles The motion trajectory factor of each point cloud data point; The acquisition The first The speed factor of each obstacle includes the following specific methods: The first The first The point cloud data point at the centroid position in the point cloud data set of the obstacle is The first The Euclidean distance between the point cloud data points at the centroid position in the point cloud data set of the obstacle is taken as the The first The speed factor of each obstacle; The specific method of obtaining the interaction factor between each obstacle and the loading and unloading vehicle at each acquisition moment according to the relative position difference between each obstacle and the loading and unloading vehicle at each acquisition moment is as follows: According to the position difference between each obstacle and the point cloud dataset of the loading and unloading vehicle at each acquisition time, the first The first The distance position difference factor of each obstacle; According to the movement direction of the point cloud data point, get the The first The point cloud dataset of obstacles The motion direction difference factor of each point cloud data point; Get the The first The calculation method of the interaction factor between obstacles and loading and unloading vehicles is: In the formula, Indicates The first The interaction factor between obstacles and loading and unloading vehicles; Indicates The first The distance position difference factor of each obstacle; Indicates The first The mean of the motion direction difference factors of all point cloud data points in the point cloud data set of obstacles; represents an exponential function with a natural constant as base; The acquisition The first The distance position difference factor of each obstacle includes the following specific methods: The first The first The point cloud data point at the centroid position in the point cloud data set of the obstacle is The Euclidean distance between the point cloud data points at the centroid position in the point cloud data set of the loading and unloading truck at the collection time is recorded as The first The first distance of the obstacle; The first The first distance to the obstacle and the The first The difference of the first distance of the obstacles is taken as the The first The distance position difference value of the obstacle; if the distance position difference value is greater than or equal to 0, 0 is used as the The first The distance position difference factor of the obstacle; if the distance position difference value is less than 0, the distance position difference value is used as the first The first Obstacle distance position difference factor; the acquisition of the The first The point cloud dataset of obstacles The motion direction difference factor of each point cloud data point includes the following specific methods: The first The first The point cloud dataset of obstacles The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The first The point cloud dataset of obstacles The movement direction value of each point cloud data point; The first The point cloud dataset of the loading and unloading truck at the time of collection is The cosine value of the acute angle between the moving direction of the point cloud data point and the horizontal direction is taken as the The point cloud dataset of the loading and unloading truck at the time of collection is The movement direction value of each point cloud data point; The first The first The point cloud dataset of obstacles The motion direction value of the point cloud data point is The point cloud dataset of the loading and unloading truck at the time of collection is The absolute value of the difference between the motion direction values ​​of the point cloud data points is taken as the The first The point cloud dataset of obstacles The motion direction difference factor of each point cloud data point.

2. The safe vehicle loading and unloading method based on laser radar according to claim 1 is characterized in that: The method of correcting the motion state factor of the obstacle according to the interaction factor between the obstacle and the loading and unloading vehicle to obtain the corrected motion state factor of each obstacle includes: The first The first The interaction factor between the obstacle and the loading and unloading vehicle is The first The product of the motion state factors of the obstacles is taken as the The first The corrected motion state factor of an obstacle.

3. The safe vehicle loading and unloading method based on laser radar according to claim 1 is characterized in that: The specific method of obtaining the risk factor of each obstacle to the loading and unloading vehicle according to the corrected motion state factor is as follows: The sequence of all the acquisition moments is recorded as the acquisition moment time series; The sequence of the corrected motion state factors of obstacles is recorded as The motion sequence of obstacles; The product of the standard deviation of the motion sequence of the obstacle and the standard deviation of the time series at the time of acquisition is recorded as the first product; The covariance between the motion sequence of each obstacle and the time series of the acquisition moment is recorded as the first covariance. The ratio of the first covariance to the first product is recorded as the first The danger factor of an obstacle to loading and unloading vehicles.

4. A safe loading and unloading vehicle system based on laser radar, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the laser radar-based safe loading and unloading method as described in any one of claims 1-3 are implemented.

Images