Automatic obstacle avoidance method and system based on Internet of Things

Abstract

The invention relates to the technical field of data processing, in particular to an automatic obstacle avoidance method and system based on the Internet of Things. The method comprises the following steps: carrying out region division on equipment to form a plurality of component regions; for each component area, determining an obstacle influencing the component area as an influencing object, and calculating the threat degree of the influencing object to the component area; for each component region, obtaining a repulsive force adjustment weight based on the threat degree of each influence object to the component region; the initial repulsive force of the influencing object is obtained, the product of the repulsive force adjusting weight and the initial repulsive force of the influencing object serves as local optimal repulsive force, and the comprehensive repulsive force of all the component areas is obtained based on the multiple local optimal repulsive force; and performing automatic obstacle avoidance by using an artificial potential energy method based on the comprehensive repulsive force and the gravitational field suffered by each component area. The method has the effects of improving the obstacle avoidance precision and reducing the collision risk.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and in particular, to an automatic obstacle avoidance method and system based on the Internet of Things. Background Art

[0002] With the development of industrial machinery, integration has become a common understanding of the equipment requirements in the chemical industry and the fine chemical industry. And skid-mounted is one of the ways of integration. Skid-mounted mainly refers to a group of equipment fixed on a chassis, which can be moved by a crowbar, facilitating the relocation of skid-mounted equipment. The traditional relocation method of skid-mounted equipment mainly relies on manual labor for movement. However, with the development of industrial machinery, the automatic transportation methods of lifting equipment or other transportation methods (such as movable chassis) have gradually replaced manual labor. Although the methods of using lifting equipment or movable chassis can replace manual handling of equipment, manual operation is still required to avoid collisions with other equipment or objects during the movement of the equipment. Especially for some chemical equipment manufacturers, special attention needs to be paid during the process of equipment storage or retrieval to avoid damage to the equipment. With the development of Internet of Things technology and electronic information technology, some automatic obstacle avoidance methods have gradually been applied to the movement process of skid-mounted equipment to cope with the complex and changeable environment, and at the same time, it can further reduce the labor burden of workers. The artificial potential field method is a common obstacle avoidance method in the field of obstacle avoidance. Its core idea is to construct a repulsive potential field around the obstacle and a gravitational potential field at the position of the target point. Based on the gravitational and repulsive forces acting on the controlled object (i.e., the moving object, which refers to the skid-mounted equipment here) in the composite field, a collision-free path is searched to guide the movement of the controlled object and achieve the automatic obstacle avoidance of the controlled object.

[0003] In the actual production environment, on the one hand, there are a wide variety of skid-mounted equipment, and their shapes and volumes are highly irregular. At the same time, during the movement of skid-mounted equipment, the obstacles in its surrounding environment are not only single static obstacles, but mostly a mixed scenario of dynamic and static obstacles. The traditional artificial potential field method regards the equipment and obstacles as point masses, and only considers the distance between the overall equipment and the obstacles during the obstacle avoidance process, unable to perceive the positions of the specific components of the equipment, resulting in the inability to identify the collision risks of specific parts of the equipment, and further leading to the situation of ineffective obstacle avoidance or untimely obstacle avoidance. Summary of the Invention

[0004] To solve the problem of ineffective or untimely obstacle avoidance in traditional obstacle avoidance methods, this application provides an automatic obstacle avoidance method and system based on the Internet of Things.

[0005] In the first aspect, this application provides an automatic obstacle avoidance method based on the Internet of Things, adopting the following technical solutions: An automatic obstacle avoidance method based on the Internet of Things, comprising the steps of: dividing the device into regions to form multiple component regions; for each component region, determining the obstacles that affect the component region as influencing objects, and calculating the threat degree of the influencing objects to the component region; for each component region, obtaining a repulsive force adjustment weight based on the threat degree of each influencing object to the component region; obtaining the initial repulsive force of the influencing object, taking the product of the repulsive force adjustment weight and the initial repulsive force of the influencing object as the local optimal repulsive force, and obtaining the comprehensive repulsive force of each component region based on the multiple local optimal repulsive forces; and performing automatic obstacle avoidance on each component region using the artificial potential energy method based on the comprehensive repulsive force and the gravitational field received by each component region. Among them, the step of calculating the threat degree of the influencing object to the component region includes: obtaining the local density of the influencing object according to the distribution of the influencing object; obtaining the motion consistency between the influencing object and the component region according to the motion state of the influencing object; and taking the ratio of the local density of the influencing object to the motion consistency as the threat degree of the influencing object to the component region.

[0006] On the one hand, in this application, the skid-mounted device is divided into multiple component regions, the comprehensive repulsive force received by each component region is calculated, and obstacle avoidance is performed on each component region based on the comprehensive repulsive force corresponding to each component region and the gravity in the artificial potential field method. Compared with the traditional artificial potential field method for obstacle avoidance, in this application, each component region can be adjusted independently during the movement of the skid-mounted device, so that the accuracy of obstacle avoidance is higher, the timeliness of obstacle avoidance is improved, and the risk of collision is reduced.

[0007] On the other hand, in this method, the influence degree of different influencing objects on the component region is calculated. For the influencing object with poor motion consistency and large local density, it is determined that the threat degree of the influencing object to the component region is relatively large. The threat degree represents the risk of collision between the influencing object and the component region. The initial repulsive force is adjusted based on the threat degree, and then the adjustment of the comprehensive repulsive force is realized. By combining the comprehensive repulsive force with the gravitational field in the artificial potential field method, the direction of movement of the component region can be changed, and accurate obstacle avoidance of the component region can be realized.

[0008] Optionally, the step of obtaining the local density of the influencing object includes: determining the reference region of the influencing object, obtaining the average value of the Euclidean distances between the influencing object and each obstacle in the corresponding reference region as the distribution distance, determining the local density of the influencing object based on the distribution distance, and the local density of the influencing object is inversely proportional to the distribution distance.

[0009] Determine the reference region of the influencing object, calculate the local density of the region where the influencing object is located based on the distance between the obstacle and the influencing object in the reference region. If the distance between the obstacle and the influencing object in the reference region is small, it indicates that the corresponding local density of the influencing object is large, which means that the difficulty of the component region in the obstacle avoidance process is higher, and thus there is also a higher collision risk.

[0010] Optionally, when both the influencing object and the obstacles in the influencing object area are in a stationary state: the motion consistency between the influencing object and the component area is 1; When there are obstacles in a moving state in the influencing object or the reference area corresponding to the influencing object, obtain the horizontal speed difference, vertical speed difference, and speed direction difference between the influencing object and the component area. The product of the horizontal speed difference, vertical speed difference, and speed direction difference is used as the total difference to calculate the motion consistency.

[0011] During the obstacle avoidance process, if both the influencing object and the obstacles in the corresponding reference area are stationary, the difficulty of obstacle avoidance in this state is relatively low. Therefore, the threat to the component area in this case is relatively small. Thus, directly define the motion consistency of the influencing object in this state as 1. When the influencing object or the obstacles in the reference area corresponding to the influencing object are moving, obtain the motion consistency between the influencing object and the component area. If the motion consistency is poor, it indicates that the motion of the influencing object and the obstacles in the corresponding reference area is chaotic, and the difficulty of obstacle avoidance is greater. Therefore, the threat level to the component area is also greater.

[0012] Optionally, the steps to determine the reference area of the influencing object include: obtaining the centroid of the obstacle, and constructing a circular area with the centroid of the obstacle as the center and a radius of as the reference area of the influencing object.

[0013] Optionally, the horizontal component velocities of the component area, the influencing object, and the obstacles in the reference area corresponding to the influencing object form a horizontal velocity sequence; the standard deviation of the horizontal velocity sequence is used as the horizontal speed difference; The vertical component velocities of the component area, the influencing object, and the obstacles in the reference area corresponding to the influencing object form a vertical velocity sequence; the standard deviation of the vertical velocity sequence is used as the vertical speed difference; Obtain the motion directions of the component area, the influencing object, and the obstacles in the reference area of the influencing object, obtain the angles between the velocity direction corresponding to the component area and the influencing object and the obstacles, form an angle sequence, and use the standard deviation of the angle sequence as the speed direction difference.

[0014] To facilitate the comparison of the differences in the motion speeds of the component area, the influencing object, and the obstacles in the reference area, decompose the motion speed and compare the differences in the horizontal and vertical component velocities respectively. The standard deviation indicates the fluctuation of the sequence, so it can reflect the differences in the horizontal component velocity, vertical component velocity, and speed direction.

[0015] Optionally, obtain the range dynamic adjustment coefficient; adjust the influence range of the influencing object based on the range dynamic adjustment coefficient, optimize the repulsive force function in the artificial potential field method; obtain the initial repulsive force of each obstacle based on the repulsive force function.

[0016] In the artificial potential field method, there is an influence range affected by obstacles, which is preset by the staff. However, for different obstacles, their own motion conditions are different. In order to reduce the risk of collision between fast-moving obstacles and the component area, the preset influence range should be adjusted to improve the obstacle avoidance accuracy.

[0017] Optionally, obtain the minimum distance from the component area to the motion path of the obstacle and the minimum time to reach the motion path of the obstacle; take the product of the minimum distance and the minimum time as the first adjustment value; take the relative velocity between the component area and the motion of the obstacle as the second adjustment value, and take the normalized result of the ratio of the second adjustment value to the first adjustment value as the range dynamic adjustment coefficient.

[0018] The dynamic adjustment coefficient changes with the relative velocity between the obstacle and the component area. For obstacles with a relatively large relative velocity, the influence range should be expanded so that the obstacle can act on the component area in advance, enabling the component area to react in advance and further reducing the risk of collision.

[0019] Optionally, the steps of adjusting the influence range of the obstacle based on the range dynamic adjustment coefficient include: for each obstacle, take the product of the preset initial range of the obstacle and the dynamic influence coefficient as the adjusted range, and take the sum of the initial range and the adjusted range as the influence range of the obstacle.

[0020] Optionally, the steps of obtaining the comprehensive repulsive force of each component area based on multiple local optimal repulsive forces include: for each component area, obtain the resultant force direction of the initial repulsive forces of all obstacles corresponding to this component area; for each obstacle, take the angle between the initial repulsive force direction of this obstacle and the resultant force direction as the direction angle, take the product of the optimal repulsive force and the cosine value of the direction angle as the local repulsive force, and take the sum of multiple local repulsive forces as the comprehensive repulsive force.

[0021] In a second aspect, the present application provides an automatic obstacle avoidance system based on the Internet of Things, adopting the following technical solution: An automatic obstacle avoidance system based on the Internet of Things includes: a processor and a memory, and the memory stores computer program instructions, which when executed by the processor implement an automatic obstacle avoidance method based on the Internet of Things as described above.

[0022] Generate a computer program for the automatic obstacle avoidance method based on the Internet of Things as described above and store it in the memory to be loaded and executed by the processor. Thus, a system is made according to the memory and the processor, which is convenient to use.

[0023] The present application has the following technical effects: Divide the skid-mounted equipment into multiple component areas, obtain the comprehensive repulsive force based on the multiple component areas, and avoid obstacles based on the comprehensive repulsive force of the multiple component areas, so as to improve the accuracy and timeliness of obstacle avoidance and reduce the risk of collision during the movement of the skid-mounted equipment. Description of the Drawings

[0024] Figure 1 It is a flowchart of a method for automatic obstacle avoidance based on the Internet of Things according to an embodiment of the present application.

[0025] Figure 2 It is a flowchart of step S2 in a method for automatic obstacle avoidance based on the Internet of Things according to an embodiment of the present application. Detailed Embodiments

[0026] An embodiment of the present application discloses a method for automatic obstacle avoidance based on the Internet of Things, which divides an irregular skid-mounted equipment into multiple component areas; determines the threat level of each influencing object to the component area based on the distribution of the influencing objects that have an impact on itself and the motion state of the influencing objects corresponding to the component area, and obtains the repulsive force adjustment weight based on the threat level of each influencing object to the component area and adjusts the initial repulsive force to obtain the local optimal repulsive force. The total repulsive force received by the component area is obtained by synthesizing the repulsive forces of multiple influencing objects on the component area. In this method, for an influencing object with a greater threat level, the final obtained optimal local repulsive force is greater. Therefore, during the obstacle avoidance process, this type of influencing object can be avoided in advance and in a timely manner, and the influence range of this type of influencing object is larger, that is, during the obstacle avoidance process, the safety distance between the component area and the influencing object is kept larger, further reducing the risk of ineffective obstacle avoidance or untimely obstacle avoidance.

[0027] Refer to Figure 1 , a method for automatic obstacle avoidance based on the Internet of Things includes steps S1 - S5.

[0028] S1: Divide the equipment into areas to form multiple component areas.

[0029] Since the movement of the skid-mounted equipment is usually horizontal, here it is divided based on the projection of the skid-mounted equipment on the horizontal plane. During the division process, each component of the skid-mounted equipment is used as a component area. Of course, in other embodiments, it can also be evenly divided based on the projection of the skid-mounted equipment on the horizontal plane to form multiple component areas.

[0030] S2: For each component area, determine the obstacles that have an impact on the component area as influencing objects, and calculate the threat level of the influencing objects to the component area.

[0031] Among them, the steps of calculating the threat level of the influencing object to the component area include: obtaining the local density of the influencing object according to the distribution of the influencing object; obtaining the motion consistency between the influencing object and the component area according to the motion state of the influencing object; and taking the ratio of the local density of the influencing object to the motion consistency as the threat level of the influencing object to the component area.

[0032] UWB (Ultra Wide Band) and IMU (Inertial Measurement Unit) are set in each component area of the skid-mounted equipment; the position of the centroid of each component area is determined by UWB, and the linear velocity of the centroid of each component area is collected by IMU. At the same time, radars are set in each component area to obtain the positions and speeds of obstacles within the moving range of each component area.

[0033] For the moving scenarios of relatively complex skid-mounted equipment, the moving scenarios usually include dynamic and static obstacles. In the artificial potential field method, for the obstacles during the movement of the skid-mounted equipment, there is a preset influence range, which is a preset initial range, indicating that when the component area falls within the initial range, the obstacle has a repulsive influence on the component area. In a region, the component area may fall within the initial ranges of multiple obstacles. For the convenience of description, in this application, such obstacles are defined as influencing objects. When the component area is at a certain position, it may fall within the initial ranges of multiple influencing objects at the same time. Therefore, one component area corresponds to multiple influencing objects. For an influencing object, if the distribution of obstacles in the area where the influencing object is located is relatively dense, then the risk of collision in this area is greater. Similarly, if the movement of an obstacle is relatively chaotic, the risk of collision with the skid-mounted equipment is also greater.

[0034] Refer to Figure 2 , step S2 includes steps S21 - S23.

[0035] S21: Obtain the local density of the influencing object according to the distribution of the influencing object.

[0036] Determine the reference area of the influencing object, obtain the average value of the Euclidean distances between the influencing object and each obstacle in the corresponding reference area as the distribution distance, and determine the local density of the influencing object based on the distribution distance. The local density of the influencing object is inversely proportional to the distribution distance.

[0037] Taking the centroid of the influencing object as the center, construct a circular area with a radius of , and use this area as the reference area. In the reference area, if the distance between the influencing object and other obstacles is smaller, it means that the density of the obstacle distribution in the reference area is greater, and thus the local density of the area where the influencing object is located is greater.

[0038] Specifically, the calculation formula of local density can be expressed as: ; where represents the local density of the influencing object ; represents the th influencing object and the th obstacle in the reference area (the obstacle includes the influencing object. When analyzing the local density of an influencing object, there may be other influencing objects in its reference area, and here other influencing objects are still regarded as obstacles), and the Euclidean distance between them; represents the number of obstacles in the reference area; represents the exponential function with the natural constant as the base.

[0039] In the formula, the Euclidean distance between each obstacle in the reference area and the influencing object is calculated, and the Euclidean distances are summed. If the sum of the Euclidean distances is small, it indicates that the interval between the obstacle and the influencing object is small, and further indicates that the local density of the influencing object is large.

[0040] S22: According to the motion state of the influencing object, obtain the motion consistency between the influencing object and the component area.

[0041] For obstacles, including moving obstacles and stationary obstacles. For stationary influencing objects, their states are relatively simple, and for such influencing objects, the difficulty of obstacle avoidance during the movement of the device is relatively low. While for moving influencing objects, the difficulty of avoidance is affected by the movement speed and direction of the influencing object itself. Therefore, the influencing objects should be distinguished.

[0042] When both the influencing object and the obstacles in the influencing object area are in a stationary state: the motion consistency between the influencing object and the component area is 1. Here, the stationary state means the state where the movement speed of the influencing object is 0.

[0043] When there is a moving obstacle in the influencing object or the reference area corresponding to the influencing object: obtain the horizontal speed difference, vertical speed difference, and speed direction difference between the influencing object and the component area, and the product of the horizontal speed difference, vertical speed difference, and speed direction difference is used as the total difference to calculate the motion consistency.

[0044] The steps to obtain the horizontal speed difference between the influencing object and the component area include: Construct a horizontal speed sequence: construct a plane coordinate system, decompose the movement speeds of the influencing object, the component area, and the obstacles in the reference area corresponding to the influencing object (in this embodiment, the movement speed refers to the movement speed of the centroid of the obstacle or the component area), and the horizontal speeds corresponding to the influencing object, the component area, and the obstacles in the reference area corresponding to the influencing object form a horizontal speed sequence, and the standard deviation of the horizontal speed is used as the horizontal speed difference.

[0045] Similarly to the acquisition of the horizontal speed difference, the vertical speed difference between the influencing object and the component area is acquired.

[0046] The movement directions of the obstacles in the component area, the influencing object, and the reference area of the influencing object are acquired, and the angles between the speed direction corresponding to the component area and the influencing object and the obstacles are acquired to form an angle sequence. The standard deviation of the angle sequence is used as the speed direction difference.

[0047] Then, the movement consistency is acquired based on the horizontal speed difference, the vertical speed difference, and the speed direction difference.

[0048] Specifically, the calculation formula of the movement consistency can be expressed as: ; where represents the movement consistency between the th influencing object and the corresponding component area; represents the horizontal speed difference; represents the vertical speed difference; represents the exponential function with the natural constant as the base; represents the speed direction difference.

[0049] The speeds of the influencing object and the component area are decomposed, so as to be able to compare the speed differences between the influencing object and the component area. The movement consistency between the component area and the influencing object is reflected by the speed and angle differences between the influencing object and the component area. The greater the speed and angle differences between the component area and the influencing object, the greater the difference in their operating states, and thus the poorer the movement consistency between the two.

[0050] S23: The ratio of the local density of the influencing object to the movement consistency is used as the threat degree of the influencing object to the component area.

[0051] Based on the local density of a certain influencing object and its movement consistency with the component area, the threat degree of the influencing object to a certain component area is calculated. If the local density of an influencing object is greater, it indicates that there are more obstacles in the adjacent range of the influencing object. Then, during the movement of the component area, the risk of collision between the influencing object and the component area is higher, and thus the threat degree of the influencing object to the operation of the component area is greater. The movement consistency of the influencing object indicates the difference in the movement between the obstacles in the corresponding reference area of the influencing object and the component area. If the movement consistency of the influencing object is poor, it indicates that the movement of the obstacles in the corresponding reference area of the influencing object is chaotic. Then, during the movement of the component area, the risk of collision with the influencing object and the obstacles in the corresponding reference area of the influencing object is higher, and thus the threat degree of the influencing object to the component area is also greater. The two act synergistically to improve the accuracy and robustness of the threat degree calculation for the threat degree of the influencing object to the component area.

[0052] S3: For each component area, obtain the repulsive force adjustment weight based on the threat level of each influencing object to the component area.

[0053] When the skid-mounted equipment moves to a certain area, each component area is respectively affected by multiple influencing objects, and different influencing objects have different threat levels to the component area. For the adjustment weight corresponding to each influencing object, obtain the sum of the threat levels of all influencing objects to the component area, take the sum of the threat levels as the total threat level, take the ratio of the threat level corresponding to each influencing object to the total threat level as the adjustment degree, and take the sum of the adjustment degree and the preset base value as the adjustment weight of the influencing object. In this embodiment, the base value is 1.

[0054] S4: Obtain the initial repulsive force of the influencing object, take the product of the repulsive force adjustment weight and the initial repulsive force of the influencing object as the local optimal repulsive force, and obtain the comprehensive repulsive force of each component area based on multiple local optimal repulsive forces.

[0055] Obtain the range dynamic adjustment coefficient: Obtain the minimum distance from the component area to the movement path of the influencing object and the minimum time to reach the movement path of the influencing object; take the product of the minimum distance and the minimum time as the first adjustment value; take the relative speed of the component area and the movement of the influencing object as the second adjustment value, and take the normalized result of the ratio of the second adjustment value to the first adjustment value as the range dynamic adjustment coefficient.

[0056] Specifically, the calculation formula of the range dynamic adjustment coefficient can be expressed as: ; In the formula, represents the influencing object range dynamic adjustment coefficient; represents the influencing object relative speed with the component area; the minimum time for the component area to reach the movement path of the influencing object; the minimum distance for the component area to reach the movement path of the influencing object. Here, the minimum distance can be understood as the distance of the perpendicular line from the component area to the movement trajectory line of the influencing object; represents the time of a preset distance, which is mainly used to eliminate the dimension of the final range dynamic adjustment coefficient and normalize the final calculation result.

[0057] Adjust the influence range of the influencing object based on the range dynamic adjustment coefficient, and optimize the repulsive force function in the artificial potential field method; obtain the initial repulsive force of each obstacle based on the repulsive force function.

[0058] For each influencing object, take the product of the preset initial range of the influencing object and the dynamic influence coefficient as the adjustment range, and take the sum of the initial range and the adjustment range as the influence range of the influencing object.

[0059] represents the first adjustment value, represents the second adjustment value, and the two adjustment values adjust the initial range in which a preset influencing object affects the component area. By adjusting the preset initial range with the first adjustment value and the second adjustment value, the effect of adjusting the repulsive force of the influencing object on the component area can be achieved. When the first adjustment value is small, it indicates that the minimum distance and the minimum time are small, indicating that the risk of collision between the corresponding influencing object and the component area is greater. Therefore, the initial range is expanded, enabling the influencing object to affect the component area in advance, thereby improving the timeliness of obstacle avoidance. Similarly, if the relative speed between the influencing object and the component area is greater, it means that the two are approaching at a faster speed, and the risk of collision is higher, so the initial range needs to be expanded.

[0060] Optimize the repulsive force function in the artificial potential field method, and substitute the optimized influence range of the influencing object into the repulsive force function in the artificial potential field method.

[0061] The formula of the optimized repulsive force function can be expressed as: ; in the formula, represents the repulsive force of the influencing object on the component area; is the proportionality coefficient; represents the influencing object and the Euclidean distance between the centroid of the component area; represents the initial range; represents the influencing object range dynamic adjustment coefficient.

[0062] represents the optimized influence range of the initial range. Applying this influence range to the repulsive force function realizes the optimization of the repulsive force function. The repulsive force function is a conventional technical means in this field and will not be elaborated here.

[0063] For each influencing object of the component area, substitute the Euclidean distance between the influencing object and the component area and the corresponding influence range into the repulsive force function, so as to obtain the initial repulsive force.

[0064] Take the product of the repulsive force adjustment weight and the initial repulsive force of the influencing object as the local optimal repulsive force, and obtain the comprehensive repulsive force of each component area based on multiple local optimal repulsive forces; ; in the formula, represents the comprehensive repulsive force received by the component area; the threat level of the influencing object to the component area; represents the number of influencing objects corresponding to the component area; represents the influencing object Initial repulsive force on the component area; Represents an influencing object The angle between the direction of the initial repulsive force of the influencing object and the direction of the resultant force of the initial repulsive forces of multiple influencing objects; Represents a base value, which is 1 in this embodiment.

[0065] S5: Based on the comprehensive repulsive force and gravitational field received by each component area, use the artificial potential energy method for automatic obstacle avoidance.

[0066] Based on the comprehensive repulsive force and gravitational field received by each component area, use the artificial potential energy method to avoid obstacles for each component area. The artificial potential field method is a conventional technical means in the art and will not be elaborated here. It constructs the movement path of the component area through the gravitational and repulsive forces received by the component area. By controlling the skid-mounted equipment to perform translation and torsion, the individual position adjustment of multiple component areas is realized, the timeliness of obstacle avoidance is improved, and the collision risk is reduced.

[0067] The embodiment of the present application also discloses an automatic obstacle avoidance system based on the Internet of Things, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, an automatic obstacle avoidance method based on the Internet of Things according to the present application is realized.

[0068] The above system also includes other components well known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here.

[0069] The above are all preferred embodiments of the present application. The protection scope of the present application is not limited by this. Therefore, all equivalent changes made according to the results, shapes, and principles of the present application should be covered within the protection scope of the present application.

Claims

1. An automatic obstacle avoidance method based on the Internet of Things, characterized in that, It includes the steps of: dividing the device into regions to form multiple component regions; for each component region, determining the obstacles that affect the component region as influencing objects, and calculating the threat level of the influencing objects to the component region; for each component region, obtaining a repulsive force adjustment weight based on the threat level of each influencing object to the component region; obtaining the initial repulsive force of the influencing object, taking the product of the repulsive force adjustment weight and the initial repulsive force of the influencing object as the local optimal repulsive force, and obtaining the comprehensive repulsive force of each component region based on multiple local optimal repulsive forces; using the artificial potential field method for automatic obstacle avoidance based on the comprehensive repulsive force and the gravitational field received by each component region; Among them, the steps of calculating the threat level of the influencing object to the component region include: obtaining the local density of the influencing object according to the distribution of the influencing object; obtaining the motion consistency between the influencing object and the component region according to the motion state of the influencing object; taking the ratio of the local density of the influencing object to the motion consistency as the threat level of the influencing object to the component region.

2. The automatic obstacle avoidance method based on the Internet of Things according to claim 1, characterized in that The steps of obtaining the local density of the influencing object include: determining the reference region of the influencing object, obtaining the average value of the Euclidean distances between the influencing object and each obstacle in the corresponding reference region as the distribution distance, and determining the local density of the influencing object based on the distribution distance. The local density of the influencing object is inversely proportional to the distribution distance.

3. An automatic obstacle avoidance method based on the Internet of Things according to claim 1, characterized in that The steps of obtaining the motion consistency between the influencing object and the component region include: When both the influencing object and the obstacles in the influencing object region are in a stationary state: the motion consistency between the influencing object and the component region is 1; When there is an obstacle in the influencing object or the corresponding reference region of the influencing object in a moving state, obtain the horizontal velocity difference, vertical velocity difference, and velocity direction difference between the influencing object and the component region. The product of the horizontal velocity difference, vertical velocity difference, and velocity direction difference is used as the total difference to calculate the motion consistency.

4. An automatic obstacle avoidance method based on the Internet of Things according to claim 2, characterized in that, The steps for determining the reference area of the influencing object include: obtaining the centroid of the obstacle, and constructing a circular area with the centroid of the obstacle as the center and a radius of as the reference area of the influencing object.

5. The automatic obstacle avoidance method based on the Internet of Things according to claim 3, characterized in that The horizontal component velocities of the component region, the influencing object, and the obstacles in the corresponding reference region of the influencing object form a horizontal velocity sequence; taking the standard deviation of the horizontal velocity sequence as the horizontal velocity difference; The vertical component velocities of the component region, the influencing object, and the obstacles in the corresponding reference region of the influencing object form a vertical velocity sequence; taking the standard deviation of the vertical velocity sequence as the vertical velocity difference; Obtain the motion directions of the component region, the influencing object, and the obstacles in the reference region of the influencing object, obtain the angles between the velocity direction corresponding to the component region and the influencing object and the obstacles, form an angle sequence, and take the standard deviation of the angle sequence as the velocity direction difference.

6. The automatic obstacle avoidance method based on the Internet of Things according to claim 1, characterized in that, The steps of obtaining the initial repulsive force of the influencing object include: obtaining a range dynamic adjustment coefficient; adjusting the influence range of the influencing object based on the range dynamic adjustment coefficient to optimize the repulsive force function in the artificial potential field method; obtaining the initial repulsive force of each obstacle based on the repulsive force function.

7. The automatic obstacle avoidance method based on the Internet of Things according to claim 6, characterized in that, The steps of obtaining the range dynamic adjustment coefficient include: obtaining the minimum distance from the component region to the motion path of the influencing object and the minimum time to reach the motion path of the influencing object; taking the product of the minimum distance and the minimum time as the first adjustment value; taking the relative velocity of the component region and the motion of the influencing object as the second adjustment value, and taking the normalized result of the ratio of the second adjustment value to the first adjustment value as the range dynamic adjustment coefficient.

8. The automatic obstacle avoidance method based on the Internet of Things according to claim 7, characterized in that The steps of dynamically adjusting the influence range of an influencing object based on a range dynamic adjustment coefficient include: for each influencing object, taking the product of the preset initial range of the influencing object and the dynamic influence coefficient as the adjustment range, and taking the sum of the initial range and the adjustment range as the influence range of the influencing object.

9. The automatic obstacle avoidance method based on the Internet of Things according to claim 8, characterized in that, The steps of obtaining the comprehensive repulsive force of each component area based on multiple local optimal repulsive forces include: for each component area, obtaining the resultant force direction of the initial repulsive forces of all the influencing objects corresponding to the component area; for each influencing object, taking the angle between the initial repulsive force direction of the influencing object and the resultant force direction as the direction angle, taking the product of the optimal repulsive force and the cosine value of the direction angle as the local repulsive force, and taking the sum of multiple local repulsive forces as the comprehensive repulsive force.

10. An automatic obstacle avoidance system based on the Internet of Things, characterized in that, including: a processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement an automatic obstacle avoidance method based on the Internet of Things according to any one of claims 1-9.

Images