Obstacle avoidance method for submarine cable inspection AUV based on sonar and fuzzy artificial potential field method

Abstract

The application discloses a submarine cable inspection AUV obstacle avoidance method based on a sonar and a fuzzy artificial potential field method. The method obtains sound wave data of the obstacle avoidance sonar through a sonar data acquisition and preprocessing system, pre-processes and analyzes the sound wave data, obtains obstacle information, and takes the minimum distance d f from the AUV to the front obstacle, the distance d g from the AUV to the target position, and the absolute value a s of the AUV turning angle as inputs of the fuzzy algorithm, and obtains the AUV speed v r by using fuzzy rules. Then, the AUV speed v r generated according to the obstacle information, the target position and the fuzzy rules is introduced into the influence factor of the sea current, the AUV is subjected to vector synthesis, an artificial potential field is generated, the AUV turning angle is obtained according to the potential field resultant force, and the path of the AUV is planned. Corresponding actions are taken in terms of the AUV turning and speed, so as to guide the mobile robot to reach the target, and an obstacle avoidance strategy is formulated to realize accurate obstacle avoidance of the AUV.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of local obstacle avoidance of AUV, and relates to a submarine cable inspection AUV obstacle avoidance method based on a sonar and a fuzzy artificial potential field method. BACKGROUND

[0002] With the continuous development of the marine economy, submarine pipelines play an increasingly important role and undertake the role of transporting oil and gas, power and communication. However, the submarine pipelines may be damaged due to external influences such as anchors and submarine organisms, and changes in the submarine environment may also cause the pipelines to be exposed and leak, thus posing certain risks, and therefore the pipelines need to be positioned and fault detected and repaired in a timely manner. It is not realistic to rely on manpower to maintain such a large pipeline system, and the risk coefficient is high.

[0003] An autonomous underwater vehicle (AUV) has the advantages of wide range of activities, low cost, and small environmental impact, and is very suitable for undertaking the task of submarine pipeline inspection. Compared with manned underwater vehicles, the AUV has the advantages of high maneuverability, no risk of personnel casualties, high adaptability and survivability, and low manufacturing and maintenance cost, and has attracted the attention of scholars from various countries and become an important device for marine exploration and scientific research.

[0004] When an AUV performs a survey task, it needs to respond to complex and variable underwater environments in real time so as to smoothly plan a path to the destination, and therefore underwater obstacle avoidance algorithms are one of the key research directions in the field of AUV control. Obstacle avoidance is to enable a mobile robot to complete the action of avoiding obstacles in the fastest time and at the lowest cost when a sensor on the mobile robot detects that obstacles appear around, and to continue to move forward according to the route after successfully avoiding the obstacles. In existing underwater obstacle avoidance algorithms for AUVs, the artificial potential field method has been widely applied in the field of AUV obstacle avoidance path planning and has gradually become a mainstream obstacle avoidance algorithm due to the advantages of simple and clear algorithm structure and fast response. However, the traditional artificial potential field method does not consider complex underwater environmental problems, and when the AUV approaches the target point, the gravitational force it receives gradually decreases, and if the target point happens to be within the influence range of the obstacle, the repulsive force it receives increases and is opposite to the direction of the gravitational force, which will cause the AUV to stop at a certain point or vibrate, resulting in the inability to reach the target point. In addition, due to the uncertain position of the obstacles in the environment, there may be a situation where the AUV receives a resultant force of 0 at some non-target points, and in this case, the AUV will stay at the local minimum point and no longer move towards the target point. SUMMARY

[0005] In view of the deficiencies of the prior art, the submarine cable inspection AUV obstacle avoidance method based on sonar and fuzzy artificial potential field method is provided, obstacle information is collected through sonar, the fuzzy algorithm and the improved artificial potential field method are combined, the AUV obstacle avoidance strategy is generated, the defects of the traditional artificial potential field method are made up, and the obstacle avoidance capability of the AUV is enhanced.

[0006] The submarine cable inspection AUV obstacle avoidance method based on sonar and fuzzy artificial potential field method specifically comprises the following steps:

[0007] Step one, the sonar is used to detect the submarine environment, the position information and the direction information of the obstacle are obtained according to echo data of the sonar, and an obstacle information map is generated.

[0008] Step two, according to the obstacle information map and the target position in step one, the minimum value d f of the AUV from the front obstacle, the minimum value d g from the target position and the absolute value a s of the AUV turning angle are obtained together input into the fuzzy algorithm, the speed v r of the AUV is obtained by using fuzzy rules. In the fuzzy controller, the number of fuzzy subsets corresponding to d f , d g and a s is 3, 2 and 6 respectively, and the number of fuzzy subsets corresponding to v r is 6.

[0009] Step three, the improved artificial potential field method is used to calculate the motion direction of the AUV, specifically comprising the following steps:

[0010] s3.1, according to the obstacle information map and the target position obtained in step one, the artificial potential field method is used to generate the attractive force field function U att (x) and the repulsive force field function U rep (x), and after negative gradient operation, the attractive force function F att (x) and the repulsive force function F rep (x) are obtained.

[0011] s3.2, according to the spatial dynamics equation and the Lagrange equation, the resultant force F tol of the AUV under the action of the attractive force field and the repulsive force field is calculated.

[0012] s3.3, in order to make the running track of the AUV comply with the sea current as much as possible, the improved artificial potential field method is used to introduce the influence F c of the sea current on the running track of the AUV, and the influence F c is defined as:

[0013] F c =K c V c

[0014] Among them, V c K represents the speed of the ocean current. c This is a constant coefficient representing the influence of ocean currents.

[0015] As a preferred option, K c The value range is 2 to 10.

[0016] With the addition of ocean currents, the net force F acting on the AUV res for:

[0017]

[0018] s3.4. Based on the AUV velocity v output by the fuzzy algorithm in step two. r And the resultant force F acting on the AUV calculated in s3.3 res Calculate the steering angle S during AUV motion. a :

[0019]

[0020] Where θ is the direction of motion of the AUV, F resy F resx The resultant force F res The components on the Y-axis and X-axis. The AUV will be in the resultant force F. res Under the action of velocity v r Deflection angle S a Proceed in the designated direction until you reach your target location.

[0021] The present invention has the following beneficial effects:

[0022] By introducing fuzzy algorithms, the obstacle avoidance performance of the algorithm is improved, and the problems of the artificial potential field method getting stuck in local minima and the target being unreachable are also solved. Considering the influence of ocean currents on obstacle avoidance path planning, the artificial potential field algorithm is improved by adding an ocean current model, so that the AUV's trajectory can adapt to the ocean current as much as possible, saving energy. The improved fuzzy artificial potential field method greatly improves the AUV's adaptability to the marine environment. Attached Figure Description

[0023] Figure 1 A flowchart of an AUV obstacle avoidance method for submarine cable inspection based on sonar and fuzzy artificial potential field method;

[0024] Figure 2 This is a schematic diagram of the sonar data acquisition and preprocessing system used in the embodiment;

[0025] Figure 3 This is an obstacle information map generated in the embodiment;

[0026] Figure 4The fuzzy rule base established in the embodiment;

[0027] Figure 5 AUV force schematic diagram for improving artificial potential field. DETAILED DESCRIPTION

[0028] The application will be further explained in connection with the accompanying drawings;

[0029] The obstacle avoidance method of the submarine cable inspection AUV based on the sonar and the fuzzy artificial potential field method obtains the sound wave data of the obstacle avoidance sonar through the sonar data acquisition and preprocessing system, pre-processes and analyzes the sound wave data, obtains the obstacle information, and according to the obstacle information map and the target position, the minimum distance d f from the AUV to the front obstacle, the distance d g from the AUV to the target position, and the absolute value a s of the turning angle of the AUV are taken as the inputs of the fuzzy algorithm, and the speed v r of the AUV is obtained by using the fuzzy rules. Then, the AUV speed v r generated according to the obstacle information, the target position and the fuzzy rules is introduced into the sea current influence factor, the AUV is vector synthesized, the artificial potential field is generated, the turning angle of the AUV is obtained according to the potential field resultant force, and the path of the AUV is planned. In terms of the turning and speed of the AUV, corresponding actions are taken, so as to guide the mobile robot to reach the target, and an obstacle avoidance strategy is formulated to realize the accurate obstacle avoidance of the AUV.

[0030] Specifically includes the following steps:

[0031] Step one, use the sonar data acquisition and preprocessing system as shown in Figure 2 to detect the submarine environment. The sonar data acquisition and preprocessing system is divided into two parts of hardware circuit and software design. The hardware circuit includes a data acquisition module and a data processing module, the data acquisition module mainly includes an STM32 processor, a power module and a communication interface; the data processing module includes a Raspberry Pi, an SD card and a communication interface. The STM32 processor in the data acquisition module sends a command to the obstacle avoidance sonar every certain period of time, and reads the return data of the obstacle avoidance sonar, each instruction can obtain 1200 sonar echo intensities on the corresponding straight line, and 360 different angle commands can obtain sonar echo intensities in 0-360° directions. The STM32 processor sends the collected sonar data to the Raspberry Pi, and the Raspberry Pi compares the sonar echo intensities. Since the stronger the signal received by the sonar at the corresponding position, the larger the data value returned, the larger the value of the sonar echo intensity, the more likely the corresponding position is an obstacle, and the position with a larger value is depicted with a deeper color to obtain an obstacle information map as shown in Figure 3 According to the color depth in Figure 3 , the position information and the direction information of the obstacle can be obtained.

[0032] Step 2: Based on the obstacle information map and target location from Step 1, obtain the minimum distance d between the AUV and the obstacle in front. f The minimum distance d from the target location g The minimum distance d from the AUV to the obstacle in front. f The distance d from the AUV to the target location g The absolute value of the AUV steering angle α s The absolute value of the turning angle α determines whether the robot is safe and the degree of safety. s When encountering large obstacles, approaching obstacles, or approaching the target location, the AUV should decelerate to improve system safety. Based on this, the AUV's speed v can be determined accordingly. r .

[0033] Establish fuzzy control rules, and d f d g a s As the input variable of the fuzzy algorithm, v r As an output variable, the fuzzy control rule is based on linguistic variables, which take values ​​of fuzzy subsets such as "large," "medium," and "small." Each fuzzy subset is represented by a membership function indicating the degree to which the precise value on the basic universe belongs to that subset. Therefore, the fuzzy control rule uses linguistic variable values ​​to replace precise values. In this embodiment, the variable is d. f d g a s Three, two, and five fuzzy subsets were constructed respectively. After fuzzification, d f ={near, middle, far}, d g ={near, far}, a s = {very small, relatively small, small, large, relatively large, very large}, v r ={very small, relatively small, small, large, relatively large, very large}. The corresponding fuzzy rule base is as follows: Figure 4 As shown, the total number of fuzzy rules is 30.

[0034] Rules 1-5 apply when the robot is moving away from obstacles and the target location; in this case, the robot should accelerate to improve task completion efficiency. Rules 26-30 apply when the robot is approaching obstacles and the target location; in this case, the robot should decelerate to avoid a safe collision with the obstacle and come to a smooth stop at the target location. Other rules can also be understood based on human experience.

[0035] Step 3: Calculate the AUV's direction of motion using the improved artificial potential field method. This includes the following steps:

[0036] s3.1. Based on the obstacle information map and target location obtained in step one, define the area around the obstacle and the area where the AUV is not desired to enter, and generate a repulsive potential field to exert a repulsive force on the AUV. The repulsive potential field function U rep (x) is:

[0037]

[0038] Where, k rep >0 indicates that the repulsive field coefficient is constant; l(x,x0) is the straight-line distance between the AUV and the obstacle, x and x0 represent the positions of the AUV and the obstacle, respectively, and l0 is the maximum influence range of the obstacle.

[0039] Repulsive potential field function U rep The graph of (x) in a two-dimensional coordinate system resembles a "high ground," with the highest repulsive potential energy at the obstacle. Performing a negative gradient operation on the repulsive potential field function yields the repulsive function F. rep (x):

[0040]

[0041] Define the target location and the region where the AUV is desired to enter, generate a gravitational potential field, and exert a gravitational force on the AUV. The gravitational potential field function is U. att (x) is:

[0042]

[0043] Where, k att >0 indicates a constant gravitational field coefficient; l(xx g x is the straight-line distance between the AUV and the target point. g Indicates the location of the target point.

[0044] Gravitational potential field function U att The graph of (x) in a two-dimensional coordinate system resembles a "valley," indicating the lowest gravitational potential energy at the target point. The gravitational function F can be obtained by performing a negative gradient calculation on the gravitational potential field function. att (x):

[0045]

[0046] s3.2, Taking a two-dimensional scene as an example, the gravitational potential field U att and repulsive potential field U rep The mathematical expression for the resultant potential field U tol For the gravitational potential field U att With repulsive potential field U rep Superposition:

[0047] U tol =U att +U rep

[0048] Based on the space dynamics equations and the Lagrange equations, calculate the resultant force F of the combined potential field on the AUV under the simultaneous action of gravitational and repulsive fields. tol :

[0049]

[0050] s3.3. When an AUV is navigating underwater, the influence of ocean currents is a significant external factor. Since AUVs carry limited energy, to ensure their trajectory follows the ocean currents as closely as possible, the artificial potential field method is improved by introducing a third force—the influence of ocean currents F—on the AUV's trajectory, distinct from the gravitational pull of the target point and the repulsive force of obstacles. c Defined as:

[0051] F c =K c V c

[0052] Among them, V c K represents the speed of the ocean current. c This is the constant coefficient for the influence of ocean currents, with a value ranging from 2 to 10.

[0053] With the addition of ocean currents, the net force F acting on the AUV res for:

[0054]

[0055] like Figure 5 As shown, in the artificial potential field space, the net force on the AUV at any position conforms to the above formula.

[0056] s3.4. Based on the AUV velocity v output by the fuzzy algorithm in step two. r And the resultant force F acting on the AUV calculated in s3.3 res Calculate the steering angle S during AUV motion. a :

[0057]

[0058] Where θ is the direction of motion of the AUV, F resy F resx The resultant force F res The components on the Y and X axes. The AUV will travel under the action of the resultant force. Each time the AUV updates its position, it will repeat the above steps to determine the minimum distance d between the AUV and the obstacle in front. f The distance d from the AUV to the target location g The absolute value of the AUV steering angle α sAs input to the fuzzy algorithm, the speed v of the AUV is calculated using fuzzy rules. r Then the speed v of the AUV r Using the improved artificial potential field method as input, the current AUV's steering angle S is calculated. a This ensures that the AUV avoids static obstacles while following the trajectory. This continues until the AUV reaches the target d. goal The distance is less than or equal to the minimum threshold distance d min This is considered as the AUV reaching the target point.

[0059] It should be understood that the exemplary embodiments described herein are illustrative and not restrictive. Although one or more embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art will understand that various changes in form and detail may be made without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method for obstacle avoidance of an AUV for submarine cable inspection based on sonar and fuzzy artificial potential field method, characterized in that: Specifically comprising the following steps: Step one, using sonar to detect the seabed environment, according to the echo data of sonar to get the position information and direction information of the obstacle, and generate the obstacle information map; Step two, according to the obstacle information map and target position in step one, get the minimum value d of AUV distance from the front obstacle f , the minimum value d of distance from the target position g , and the absolute value a of AUV turning angle s together input into the fuzzy algorithm, get the speed v of AUV by using fuzzy rules r ; the fuzzy rules are input d f ={near, medium, far}, d g ={near, far}, a s ={very small, small, small, large, large, very large}, and output v r ={very small, small, small, large, large, very large} Step three, according to the improved artificial potential field method, the moving direction of AUV is calculated, specifically comprising the following steps: Step one, using sonar to detect the seabed environment, according to the echo data of sonar to get the position information and direction information of the obstacle, and generate the obstacle information map; s3.1, according to the obstacle information map and target position obtained in step one, generate the attractive field function U att (x) and repulsive field function U rep (x) and repulsive field function F att (x) and repulsive field function F rep (x) and repulsive field function F s3.2, according to the spatial dynamics equation and the Lagrange equation, the resultant force F of the AUV under the action of the synthetic potential field of the attractive field and the repulsive field is calculated tol ; s3.3, the improved artificial potential field method, the influence of the sea current on the AUV travel trajectory F is introduced c , defined as: F c = K c V c wherein V c is the speed of the sea current; K c is the sea current influence constant coefficient; After adding the current factor, the resultant force F on the AUV res is: s3.4, the AUV velocity v output by the fuzzy algorithm in step two r and the resultant force F calculated in s3.3 res , the steering angle S in the AUV motion is calculated a : where θ is the moving direction of the AUV, F resy , F resx are the components of the resultant force F res in the Y-axis and X-axis respectively; the AUV moves in the direction of the deflection angle S res at a speed v r under the action of the resultant force F a until it reaches the target point.

2. The method of claim 1, wherein the method further comprises: A repulsive potential field is generated around the obstacles and the areas where the AUV is not desired to enter, and a repulsive force is generated to the AUV, the repulsive potential field function U rep (x) is: where k rep > 0 is the repulsive force field coefficient constant; l(x, x0) is the straight-line distance between the AUV and the obstacle, x and x0 represent the positions of the AUV and the obstacle, respectively, and l0 is the maximum influence range of the obstacle; the repulsive force function F rep (x) is obtained by performing a negative gradient operation on the repulsive potential field function.

3. The method of claim 1, wherein the method further comprises: The target position and the area where the AUV is expected to enter are defined to generate a gravitational potential field, and a gravity is generated on the AUV, the gravitational potential field function U att (x) is: where k att > 0 is the gravitational field coefficient constant; l(x-x g ) is the straight-line distance of the AUV from the target point, x g represents the target point position; and the negative gradient of the gravitational potential field function is calculated to obtain the gravitational function F att (x):

4. The method of claim 1, wherein the method further comprises: K c the value range of n is 2-10.

Images