Multi-heterogeneous unmanned equipment ship surface path planning method for improving artificial potential field method

Abstract

The invention discloses a multi-heterogeneous unmanned equipment ship surface path planning method for improving an artificial potential field method, and the method comprises the steps: initializing the field environment information of multi-heterogeneous unmanned equipment, and determining the starting point coordinate, the obstacle coordinate, the target point coordinate, the obstacle maximum influence radius and the step length information of the multi-heterogeneous unmanned equipment; setting a potential field model, adjusting the action of repulsive force and gravitational force in real time by introducing a dynamic factor coefficient, performing obstacle avoidance regulation and control on the multi-heterogeneous unmanned equipment in operation, controlling the multi-heterogeneous unmanned equipment to operate towards a target point through a gravitational function, and adding a longitudinal random factor into the gravitational function to dynamically adjust the direction and strength of the gravitational force; and performing path optimization on the multi-heterogeneous unmanned equipment by adopting an improved artificial potential field method, and controlling the multi-heterogeneous unmanned equipment to run towards a target point.

Description

Technical Field

[0001] The invention belongs to the field of aviation support technology and intelligent unmanned system, and proposes a multi-heterogeneous unmanned equipment deck path planning method using an improved artificial potential field method. Background Art

[0002] The deck of an aircraft carrier needs to undertake a large number of tasks such as take-off and landing, maintenance, and replenishment of carrier-based aircraft. The efficiency and accuracy of these tasks directly affect the overall combat capability of the fleet. In this context, in order to improve the operational efficiency of the deck and reduce personnel losses, unmanned and intelligent technologies are gradually applied to the operations of the aircraft carrier deck. When multiple heterogeneous unmanned equipment (such as unmanned carrier-based aircraft and unmanned tractors) move in coordination, it is necessary not only to consider the plane position, but also to comprehensively consider the height factors of multiple heterogeneous unmanned equipment, and reasonably use the height difference to improve the utilization rate of limited space resources on the deck. When unmanned carrier-based aircraft and unmanned tractors operate on the deck, different equipment has different heights and volumes, resulting in the inability of traditional two-dimensional plane path planning to solve the conflict problem in a three-dimensional dynamic environment. Therefore, for the coordinated movement of multiple heterogeneous unmanned equipment in a complex dynamic environment, formulating an efficient path planning method has become a key issue that needs to be urgently solved in the field of aviation support technology and intelligent unmanned systems.

[0003] The deck of an aircraft carrier is a highly dynamic and high-operation-density environment. The frequent take-off and landing, movement and parking of carrier-based aircraft place extremely high demands on path planning. In addition, various ground auxiliary equipment such as unmanned tractors, unmanned refueling vehicles, unmanned maintenance vehicles, etc. need to coordinate operations with unmanned carrier-based aircraft and work synchronously and closely on the deck. Due to the significant differences in the size, motion characteristics and functions of these unmanned equipment, the complexity of their collaborative operations also increases. For example, unmanned tractors are large in size and move relatively slowly, while unmanned carrier-based aircraft require higher maneuverability and flexibility. Therefore, the aircraft carrier deck performs high-density collaborative operations of multiple equipment during peak combat hours. How to effectively coordinate the three-dimensional path planning of multiple heterogeneous unmanned equipment on the deck and effectively solve the problem of plane overlap, improve equipment utilization efficiency, avoid collisions and improve operational efficiency has become a key technical issue in the operation of modern aircraft carrier decks.

[0004] Traditional operations on aircraft carrier decks usually rely on manual command and real-time monitoring to coordinate the movement of carrier-based aircraft and ground auxiliary equipment. This mode of operation is inefficient and prone to accidents due to manual operation errors. In addition, with the popularization of unmanned systems in future operations, the traditional manual operation mode obviously cannot meet the higher requirements for the efficiency and safety of aircraft carrier deck operations in future wars. Therefore, intelligent path planning technology will become the core technical means to solve the problems of aircraft carrier deck operation efficiency and safety in the future. However, the aircraft carrier deck is highly dynamic, the environment is complex, and the operation timeliness requirements are extremely high, which puts higher technical standards on the existing path planning technology.

[0005] The unmanned equipment on the deck of future aircraft carriers will be composed of a variety of heterogeneous systems. Unmanned carrier-based aircraft need to perform a series of operations such as take-off and landing, movement, parking and maintenance, while ground auxiliary equipment such as unmanned tractors and unmanned refueling trucks mainly provide necessary support for the operation of unmanned carrier-based aircraft. The collaborative needs between these multi-heterogeneous equipment determine that path planning must consider multi-objective and multi-task parallel processing to ensure efficient collaborative operation of various types of equipment on the deck. For example, when the unmanned tractor tows the unmanned carrier-based aircraft for transfer, it forms a dynamic "combination", and the path planning problem of the combination is more complex than that of a single equipment. The combination occupies a large space on the deck and needs to have flexible obstacle avoidance capabilities to avoid collisions with other unmanned equipment or carrier-based aircraft.

[0006] In this highly complex dynamic environment, traditional path planning algorithms (such as A * Although A algorithm, Djikstra algorithm, etc. can provide good path planning effects in static or low-dynamic environments, they are obviously insufficient in dealing with complex environments such as aircraft carrier decks with high dynamics, multiple obstacles, and multiple heterogeneous equipment. * The algorithm and Djikstra algorithm rely on the assumption of a static environment, which makes it difficult for them to quickly respond to frequently changing obstacles and complex interactions between multiple equipment in a dynamic environment. In addition, these algorithms usually focus on calculating the global optimal path of a single equipment, which makes it difficult to meet the requirements of real-time, coordination and efficiency in the deck operation environment. Therefore, traditional path planning algorithms are less efficient in dynamic environments, prone to suboptimal paths or reduced operating efficiency, and cannot fully guarantee the collaborative operation and dynamic obstacle avoidance requirements of multiple heterogeneous unmanned equipment.

[0007] At present, path planning technology still faces many challenges for the collaborative operation of multiple heterogeneous equipment in dynamic environments. Traditional path planning methods based on graph theory (such as A *Algorithms such as the Djikstra algorithm are relatively stable in the calculation process and have strong deterministic results, but they mostly assume that the environment is static or of low complexity. When obstacles in the environment change frequently or the types of equipment are diverse, the effectiveness of such algorithms will be significantly reduced. In addition, these algorithms are usually only applicable to path planning for a single piece of equipment and lack sufficient optimization capabilities for the coordinated movement of multiple pieces of equipment. Therefore, in order to cope with the complex and dynamic requirements of future aircraft carrier deck operations, it is necessary to introduce more advanced dynamic optimization path planning technology.

[0008] As a path planning method based on physical principles, the artificial potential field method (APF) has been widely used in the field of robot obstacle avoidance and path planning in recent years. This method guides unmanned equipment to avoid obstacles and reach the target position by simulating the attraction of the target point to the equipment and the repulsion of obstacles. Its advantages are fast calculation speed, simple principle, and suitable for path planning in dynamic environments. However, the traditional artificial potential field method has limitations in dealing with complex dynamic environments, and is prone to local minimum problems, resulting in suboptimal paths. In addition, in scenarios such as aircraft carrier decks where multiple heterogeneous unmanned equipment work together, the artificial potential field method shows certain deficiencies in dealing with dynamic obstacle avoidance of multiple targets and multiple equipment. Therefore, how to improve the artificial potential field method so that it has better adaptability and robustness in complex dynamic environments will become an important research direction for improving the efficiency of collaborative operations of unmanned equipment. Summary of the invention

[0009] In view of the problems existing in the prior art, the present invention discloses a multi-heterogeneous unmanned equipment shipboard path planning method using an improved artificial potential field method, which specifically includes the following steps:

[0010] Initialize the on-site environment information of the multi-heterogeneous unmanned equipment, determine the starting point coordinates, obstacle coordinates, target point coordinates, maximum obstacle influence radius and step length information of the multi-heterogeneous unmanned equipment;

[0011] Set up a potential field model, introduce dynamic factor coefficients to adjust the effects of repulsion and gravity in real time, perform obstacle avoidance control for multiple heterogeneous unmanned equipment during operation, control multiple heterogeneous unmanned equipment to run towards the target point through the gravity function, and add longitudinal random factors to the gravity function to dynamically adjust the gravity direction and intensity;

[0012] The improved artificial potential field method is used to optimize the paths of multiple heterogeneous unmanned equipment and control them to move towards the target point.

[0013] Furthermore, the potential field model includes a repulsive potential energy function, a repulsive force function, an attractive potential energy function and an attractive force function; the repulsive potential energy function is improved to:

[0014]

[0015] Among them, U rep ′ is the potential field function of the repulsive field, k rep To control the strength parameter of the repulsive field, R o represents the radius of the obstacle, x represents the location of the heterogeneous unmanned equipment, and x o,i represents the position of the ith obstacle, d i (x,x o,i ) is the distance from the ith obstacle to the unmanned equipment, d o is the influence range of the obstacle, α and β are the added dynamic adjustment factors, and their sizes are related to d i (x,x o,i ) is related to i (x,x o,i )≥d o When α is 1, according to the conventional gravitational function: when R o <d i (x,x o,i )<d o , heterogeneous unmanned equipment is in the gravitational transition zone, when d i (x,x o,i )≥R o +d o When d o <d i (x,x o,i )<R o +d o When d i (x,x o,i )<d o When , the speed direction change of the corresponding heterogeneous unmanned equipment increases, and due to the existence of β, the linear increase of the repulsive force will make the obstacle avoidance path smoother.

[0016] Furthermore, the gravitational potential energy function is improved as follows:

[0017]

[0018] x g represents the location of the target point, d(x,x g ) represents the distance from the heterogeneous unmanned equipment to the target point. When d(x,x g )=0, it means that the heterogeneous unmanned equipment has reached the target point, and the gravity is 0 at this time, where ε is a positive parameter and satisfies ε<d(x o,j ,x g ).

[0019] Furthermore, a longitudinal random factor is added to the gravity function to dynamically adjust the gravity direction and establish an XY coordinate system. att (X) and repulsive force f rep (X) is equal, the angle between gravity and the X-axis is obtained. Let the angle between the new gravity and the X-axis be γ, then the new gravity is:

[0020]

[0021] τ is a longitudinal random factor, and the angle δ between the new gravity and the x-axis is:

[0022]

[0023] Then when When , the new gravitational expression is:

[0024]

[0025] The size of the longitudinal random factor τ is only related to the angle δ, but it needs to satisfy:

[0026]

[0027] Due to the adoption of the above-mentioned technical scheme, the present invention provides a method for ship deck path planning for multiple heterogeneous unmanned equipment based on an improved artificial potential field method. The method designs a MATLAB simulation program for obstacle avoidance path optimization based on the improved artificial potential field method. Through the program, the path planning effect of multiple heterogeneous unmanned equipment in the complex environment of the ship deck is verified, demonstrating efficient obstacle avoidance path optimization performance and having high path planning efficiency, which provides important reference value for the practical application of intelligent aviation support.

[0028] This method significantly improves the success rate of obstacle avoidance, enables heterogeneous equipment to more calmly cope with complex deck environments during path planning, and ensures that multiple heterogeneous unmanned equipment reaches the target point safely and stably; in addition, this method can adapt to unmanned equipment of different types and sizes, such as unmanned tractors, unmanned carrier-based aircraft and various types of ship-surface equipment, thereby improving the wide adaptability and universality of path optimization strategies in practical applications; the present invention can also adapt to unmanned equipment of different types and sizes, such as unmanned tractors, unmanned carrier-based aircraft and various types of ship-surface equipment, thereby improving the wide adaptability and universality of path optimization strategies in practical applications. The present invention improves the problem that the traditional artificial potential field method is prone to falling into local minimum points, ensures that path planning can effectively avoid stagnation, and thus improves the applicability and reliability of multiple heterogeneous unmanned equipment in complex tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] 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 some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 This is a flow chart of the multi-heterogeneous unmanned equipment shipboard path planning method of improving the artificial potential field method of the present invention;

[0031] Figure 2 A flow chart of an algorithm in a path planning method in the method of the present invention;

[0032] Figure 3 Schematic diagram of the potential field action area in the method of the present invention;

[0033] Figure 4 Schematic diagram of force analysis of heterogeneous unmanned equipment under the improved artificial potential field method in the method of the present invention

[0034] Figure 5 Schematic diagram of path planning based on the traditional artificial potential field method in the method of the present invention

[0035] Figure 6 Schematic diagram of path planning for a single heterogeneous unmanned equipment on a deck in the method of the present invention

[0036] Figure 7 is the distance map between the heterogeneous unmanned equipment and obstacles in the method of the present invention

[0037] Figure 8 Schematic diagram of path planning for multiple heterogeneous unmanned equipment on the deck in the method of the present invention DETAILED DESCRIPTION

[0038] In order to make the technical solutions and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention:

[0039] like Figure 1 The multi-heterogeneous unmanned equipment shipboard path planning method of the improved artificial potential field method shown in the figure specifically comprises the following steps:

[0040] S1. Initialize environmental information, determine the starting point coordinates of the multi-heterogeneous unmanned equipment, the coordinates of the obstacles, the coordinates of the target point, the maximum impact radius of the obstacles, and the step size of the multi-heterogeneous unmanned equipment;

[0041] S2. Obtain environmental information and build a virtual force field:

[0042] S21, setting a potential field model, specifically including defining a repulsive potential field function, a repulsive force function, a gravitational potential field function and a gravitational function, to lay a foundation for path planning;

[0043] S22. Introducing dynamic factor coefficients: Introducing dynamic factor coefficients based on traditional gravity and repulsion functions to adjust the effects of repulsion and gravity in real time to ensure that multiple heterogeneous unmanned equipment avoids collisions during path planning and achieves collaborative obstacle avoidance effects;

[0044] S23. Improve the gravity function to solve the problem of unreachable targets: Further optimize the gravity function so that unmanned equipment can overcome the problem of unreachable targets encountered in the traditional artificial potential field method and improve the coverage of path planning;

[0045] S24. Introduce longitudinal random factors: Add random factors to the gravity function to dynamically adjust the gravity direction and intensity, break the local minimum trap, ensure the consistency of the path planning process, and enable the unmanned equipment to continue to move towards the target and avoid stagnation;

[0046] S3. Based on the above improvements, the improved artificial potential field method is used to find paths for multiple heterogeneous unmanned equipment, and the optimal path is obtained through multiple iterations of optimization.

[0047] S31, calculating the resultant force: at each moment, according to the current position of the unmanned equipment, calculating the resultant force from all the attraction fields and repulsion fields (refer to S2);

[0048] S32. The direction of the resultant force can be found through the gradient descent method to determine the next moving direction.

[0049] Furthermore, in step S1, initializing the environment information specifically includes: the system needs to initialize the environment information, including determining the starting point coordinates (x, y) and the target point coordinates (x g ,y g ), the coordinates of the obstacle (x o ,y o ), the radius of the obstacle R o , Maximum impact range d o The number of obstacles n and the step size of multi-heterogeneous unmanned equipment are set to 1. In addition, the gravity coefficient k needs to be set att and the repulsion coefficient k rep , to ensure the accuracy and efficiency of path planning.

[0050] In step S2, environmental information is obtained and a virtual force field is constructed.

[0051] S21, the path planning step is to first establish a potential field model. Read the initialized environment information, build a virtual force field, and set the potential field model. The potential field model includes a repulsive potential field function Urep , repulsion function f rep , gravitational potential field function U att and the gravitational function f att , laying the foundation for subsequent path planning.

[0052] The repulsive potential field function is set as:

[0053]

[0054] Among them, U rep is the potential field function of the repulsive field, R is the distance between the heterogeneous unmanned equipment and the obstacle, is the safe distance of the ith obstacle, ξ is the intensity parameter controlling the repulsive field, and n is the number of obstacles.

[0055] The gravitational potential field function is set as:

[0056]

[0057] Among them, U att is the potential field function of the gravitational field, d is the distance from the heterogeneous unmanned equipment to the target point, and ξ is the intensity parameter for controlling the gravitational field.

[0058] S22. Introducing dynamic factor coefficients: Based on the traditional gravity and repulsion functions, dynamic factor coefficients are introduced to adjust the effects of repulsion and gravity in real time to ensure that multiple heterogeneous unmanned equipment avoid collisions during path planning and achieve collaborative obstacle avoidance effects.

[0059] Since the traditional artificial potential field method has many drawbacks, a new gravitational and repulsive potential field function is proposed based on the influence range of gravity and the magnitude of gravity and repulsion. The improvement strategy is to add dynamic factor coefficients on the basis of the traditional artificial potential field function, so that heterogeneous unmanned equipment can avoid the above situation without affecting the overall performance, and make the path reach the target point more smoothly.

[0060] First, construct a new gravitational potential field function for the gravitational source:

[0061]

[0062] And the repulsive potential field function is improved as follows:

[0063]

[0064]

[0065] Among them, R o Represents the radius of the obstacle, d i is the distance from the ith obstacle to the target point, d ois the influence range of the obstacle, α and β are the added dynamic adjustment factors, and their sizes are related to d i (x,x o,i ). From the perspective of the improved gravitational potential field, when d i (x,x o,i ) is large, α is 1, according to the conventional gravitational function: when R o <d i (x,x o,i )<d o , the heterogeneous unmanned equipment is in the gravitational transition zone. Due to the change of α, the gravitational effect on the heterogeneous unmanned equipment decreases faster, the speed direction and magnitude of the heterogeneous unmanned equipment are easier to change, and obstacle avoidance is easier. For the new repulsion function, when d i (x,x o,i ) is large, the heterogeneous unmanned equipment is in the area without repulsion; when d o <d i (x,x o,i )<R o +d o When d is , the heterogeneous unmanned equipment is in the transition zone of repulsion. It can be seen that it has a wider range of influence than the traditional repulsion, and the speed direction of the heterogeneous unmanned equipment changes more. And due to the existence of β, the linear increase of repulsion will also make the obstacle avoidance path smoother; when d i (x,x o,i )<d o , according to the traditional repulsive force. The specific potential field action area is as follows Figure 3 shown.

[0066] S23. Improve the gravity function to solve the problem of unreachable targets.

[0067] The specific improvements of this gravitational potential field are as follows:

[0068]

[0069] The above formula can be simplified to:

[0070]

[0071] There is a positive parameter ε, and the selection only needs to satisfy ε<d(x o,j ,x g ), x o,j represents the coordinates of obstacles that affect the heterogeneous unmanned equipment to reach the target point. g ) = 0, it means that the heterogeneous unmanned equipment has reached the target point, and the gravity is also 0 at this time;

[0072] When heterogeneous unmanned equipment is far away from the target,

[0073]

[0074] Compared with the traditional gravitational potential field function, the improved new gravitational potential field is stronger; when heterogeneous unmanned equipment encounters the problem of not being able to reach the target, due to the addition of part of the gravity field, the combined force of the heterogeneous unmanned equipment will make it move towards the target and eventually reach the target.

[0075] S24. Add random factors to the gravity function, dynamically adjust the gravity direction and intensity, break the local minimum trap, ensure the continuity of the path planning process, and enable the unmanned equipment to continue to move towards the target and avoid stagnation. Figure 4 As shown in the figure, the gravitational force of the target point on the heterogeneous unmanned equipment and the repulsive force of the obstacle are balanced and 180° away from each other. With the addition of the longitudinal random factor, a new gravitational force breaks the balance relationship, causing the heterogeneous unmanned equipment to move toward the target.

[0076] like Figure 4 As shown in the figure, an XY coordinate system is established. When f att (X) = f rep (X), we can get the angle between gravity and the X-axis. Let the angle between the new gravity and the X-axis be δ, where the new gravity is:

[0077]

[0078] τ is a longitudinal random factor, and the angle δ between the new gravity and the x-axis is:

[0079]

[0080] Then when When , the new gravitational expression is:

[0081]

[0082] The size of the longitudinal random factor τ is only related to the angle δ, but it needs to satisfy:

[0083]

[0084] In step S3, based on the above improvements, the improved artificial potential field method is applied to find paths for multiple heterogeneous unmanned equipment to achieve safe arrival at the target point.

[0085] S31, calculate the resultant force: at each moment, according to the current position of the unmanned equipment, calculate the resultant force from all the attraction fields and repulsion fields (refer to S2).

[0086] S32. Determine the moving direction and update the position: Use the gradient descent method to accurately calculate the direction of the resultant force acting on the unmanned equipment and its corresponding velocity vector, and then iteratively update its path based on the displacement principle to ensure that each step moves in the direction where the potential energy decreases the fastest until it safely reaches the target point.

[0087] The target point exerts an attraction on the unmanned equipment, and its strength increases as the distance decreases. The gradient of the attraction of the target point on the unmanned equipment can be expressed as:

[0088]

[0089] Among them, k att is the gravitational proportionality coefficient, x is the position of the unmanned equipment, and x goal are the target point coordinates of the unmanned equipment.

[0090] Each obstacle generates a repulsive force on the unmanned equipment to prevent it from getting too close. The repulsive force usually adopts an exponential decay model, that is, the repulsive force weakens rapidly as the distance increases. For each obstacle, the gradient of the repulsive force on the unmanned equipment can be expressed as:

[0091]

[0092] Among them, k rep is the repulsive force proportionality coefficient, x o is the position coordinate of the obstacle, d(x,x o ) is from the current position x of the unmanned equipment o Distance to obstacle, d o is the influence radius of the obstacle.

[0093] Add up the gradients of all the attractive and repulsive forces to get the total gradient

[0094]

[0095] In order to ensure that the unmanned equipment can move in the direction where the potential energy decreases fastest, the gradient descent method is used to find the direction of the resultant force. The potential field moves in the direction where it decreases fastest, that is, in the negative direction. Its velocity vector v(x) can be calculated by the following formula:

[0096]

[0097] Here, η is the step size, which controls the amount of movement in each iteration.

[0098] Finally, the position of the unmanned equipment is updated according to the velocity vector v(x). This step is the actual process of moving and is also the final output of path planning:

[0099] x t+1 =x t +v(x t )·Δt

[0100] Among them, x t+1is the position of the unmanned equipment at time t+1, x t is the position of the unmanned equipment at time t, and Δt is the time interval. This update process will be repeated at each time interval until the unmanned equipment reaches the target point and stops.

[0101] In Embodiment 1, path planning for a single unmanned carrier aircraft to avoid multiple shipboard unmanned devices on the deck and path planning between multiple heterogeneous unmanned equipment (taking an unmanned carrier aircraft and an unmanned tractor as examples) were respectively carried out based on an improved artificial potential field method for path planning of multiple heterogeneous unmanned equipment on the shipboard.

[0102] Among them, the path planning for a single unmanned carrier aircraft to avoid multiple shipboard unmanned devices on the deck based on an improved artificial potential field method for path planning of multiple heterogeneous unmanned equipment on the shipboard is set as follows. Set the position of the unmanned carrier aircraft as [4m 4m 0m], the target point position as [88m 7m 0m], and set the unmanned tractor as an obstacle on the line connecting the starting point and the target point. Taking a cube with a radius of 0.5m as an example, the position in the vertical direction represents the height of the unmanned tractor. The specific obstacle information is shown in Table 1:

[0103] Table 1 Obstacle information under the path planning of a single heterogeneous unmanned equipment

[0104]

[0105] The path planning between multiple heterogeneous unmanned equipment based on an improved artificial potential field method for path planning of multiple heterogeneous unmanned equipment on the shipboard is set as follows. Set the starting point position of the No. 1 unmanned carrier aircraft as [8m 14m 0m], the starting point position of the No. 2 unmanned carrier aircraft as [18m 3m 0m], the starting point position of the No. 3 unmanned carrier aircraft as [13m 9m 0m], and the starting point position of the unmanned tractor as [15m 20m 0m]. Their target point positions are [50m 12m 0m], [40m 24m 0m], [60m 6m 0m], [45m 3m 0m] respectively. The obstacle information is shown in Table 2:

[0106] Table 2 Obstacle information under the path planning of multiple heterogeneous unmanned equipment

[0107]

[0108] Comparative example

[0109] Comparative example 1. The present invention also compared and tested the path planning simulation based on the traditional artificial potential field method. The path planning simulation of the traditional artificial potential field method is set as follows. Set the starting point position as [0m 0m] , and the target point position as [5m 5m], a cube obstacle with a radius of 1m is set on the line connecting the starting point and the target point. The coordinates of the obstacle are [1.5m,1.5m] , [3.5m,3.5m] .

[0110] Comparative Example 1 The path obtained by path planning based on the traditional artificial potential field method is as follows Figure 5 As shown in the figure, the green curve is the path obtained by path planning based on the traditional artificial potential field method, and the black cube is an obstacle. Figure 5 It can be seen from the figure that the path planning based on the traditional artificial potential field method falls into a local minimum and does not reach the specified target point. However, the difference is that Figure 6 In the figure, the black cubes are obstacles, representing some equipment on the deck (unmanned tractors or ship islands), and the red curve codes are the paths planned based on the improved artificial potential field method. Figure 6 In the effect diagram of the path planning of a single heterogeneous unmanned equipment on a deck based on the improved artificial potential field method, it can be seen that the present invention effectively avoids the drawbacks of the traditional artificial potential field method. Figure 7 In the figure, different colors represent the real-time distance display between obstacles at different locations and the target heterogeneous unmanned equipment. Figure 7 As shown in the figure, it can be seen that the shortest distance between heterogeneous unmanned equipment is greater than the safety distance of 0.5m, which effectively avoids the collision between heterogeneous unmanned equipment during the path planning process and the safety threshold can be manually modified; from the example 1 Figure 8 In the figure, black cubes represent obstacles, and other colors represent relevant information of the target heterogeneous unmanned equipment, including the starting point and the respective planned paths. Figure 8 It can be seen that the present invention reduces the local minimum problem, improves the reliability and safety of obstacle avoidance, can approach the target point more smoothly, reduces oscillation, generates a more optimized path, reduces path interference between multiple UAVs, and improves the overall performance.

[0111] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A multi-heterogeneous unmanned equipment shipboard path planning method based on an improved artificial potential field method, characterized in that include: Initialize the on-site environment information of the multi-heterogeneous unmanned equipment, determine the starting point coordinates, obstacle coordinates, target point coordinates, maximum obstacle influence radius and step length information of the multi-heterogeneous unmanned equipment; Set up a potential field model, introduce dynamic factor coefficients to adjust the effects of repulsion and gravity in real time, perform obstacle avoidance control for multiple heterogeneous unmanned equipment during operation, control multiple heterogeneous unmanned equipment to run towards the target point through the gravity function, and add longitudinal random factors to the gravity function to dynamically adjust the gravity direction and intensity; The improved artificial potential field method is used to optimize the paths of multiple heterogeneous unmanned equipment and control them to move towards the target point.

2. According to the improved artificial potential field method of claim 1, the multi-heterogeneous unmanned equipment shipboard path planning method is characterized by: The potential field model includes a repulsive potential energy function, a repulsive force function, an attractive potential energy function and an attractive force function; the repulsive potential energy function is improved to: Among them, U rep ′ is the potential field function of the repulsive field, k rep To control the strength parameter of the repulsive field, R o represents the radius of the obstacle, x represents the location of the heterogeneous unmanned equipment, and x o,i represents the position of the ith obstacle, d i (x,x o,i ) is the distance from the ith obstacle to the unmanned equipment, d o is the influence range of the obstacle, α and β are the added dynamic adjustment factors, and their sizes are related to d i (x,x o,i ) is related to i (x,x o,i )≥d o When α is 1, according to the conventional gravitational function: when R o <d i (x,x o,i )<d o , heterogeneous unmanned equipment is in the gravitational transition zone, when d i (x,x o,i )≥R o +d o When d o <d i (x,x o,i )<R o +d o When d i (x,x o,i )<d o When , the speed direction change of the corresponding heterogeneous unmanned equipment increases, and due to the existence of β, the linear increase of the repulsive force will make the obstacle avoidance path smoother.

3. The method for multi-heterogeneous unmanned equipment shipboard path planning based on an improved artificial potential field method according to claim 2, characterized in that: The gravitational potential energy function is improved as follows: x g represents the location of the target point, d(x,x g ) represents the distance from the heterogeneous unmanned equipment to the target point. When d(x,x g )=0, it means that the heterogeneous unmanned equipment has reached the target point, and the gravity is 0 at this time, where ε is a positive parameter and satisfies ε<d(x o,j ,x g ).

4. The method for multi-heterogeneous unmanned equipment shipboard path planning based on an improved artificial potential field method according to claim 2, characterized in that: Add a longitudinal random factor to the gravity function to dynamically adjust the gravity direction and establish an XY coordinate system. att (X) and repulsive force f rep (X) is equal, the angle between gravity and the X-axis is obtained. Let the angle between the new gravity and the X-axis be γ, then the new gravity is: τ is a longitudinal random factor, and the angle δ between the new gravity and the x-axis is: Then when When , the new gravitational expression is: The size of the longitudinal random factor τ is only related to the angle δ, but it needs to satisfy:

Images