Intelligent Cruise Control Method for Underwater Dredging Robots Based on Environmental Perception

Abstract

This application relates to the field of underwater dredging robot cruise control technology, specifically to an intelligent cruise control method for underwater dredging robots based on environmental perception. This method includes: determining the trajectory deviation distance by measuring the positional distance of the underwater dredging robot at each moment and all moments within a previous preset time period; determining a second gain adjustment coefficient based on the changing trend of the trajectory deviation distance at all moments within the previous preset time period, and combining this with a first gain adjustment coefficient to determine a final gain adjustment coefficient, thereby optimizing the gain coefficient of the sliding mode controller and controlling the attitude of the underwater dredging robot. This application solves the problem of lag-induced heading error in underwater dredging robots caused by the fixed gain of traditional sliding mode control. By dynamically fusing multi-source trend data to achieve adaptive gain optimization, it improves the dredging efficiency of the underwater dredging robot.

Description

Technical Field

[0001] This application relates to the field of underwater dredging robot cruise control technology, specifically to an intelligent cruise control method for underwater dredging robots based on environmental perception. Background Technology

[0002] Underwater dredging robots are intelligent devices that can replace manual labor to efficiently and safely remove silt and debris in complex and dangerous underwater environments. They are suitable for emergency dredging of rivers, drainage pipes, culverts, and underground garages in cities and towns after typhoons and rainstorms.

[0003] Underwater dredging robots are prone to heading deviations due to slippage and tilting in complex underwater environments. Existing methods often obtain the navigation path through environmental perception and use sliding mode controllers for heading correction. However, the gain coefficient of traditional sliding mode controllers is mostly a fixed value set empirically, which is difficult to cope with the hysteresis heading errors caused by internal factors such as power supply fluctuations and rotation speed fluctuations, as well as external factors such as uneven silt distribution and water flow disturbances. This causes the robot to deviate from the predetermined path, reducing the dredging efficiency of the underwater dredging robot. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an intelligent cruise control method for underwater dredging robots based on environmental perception, thereby resolving the existing issues.

[0005] The intelligent cruise control method for underwater dredging robots based on environmental perception in this application adopts the following technical solution:

[0006] One embodiment of this application provides an intelligent cruise control method for an underwater dredging robot based on environmental perception, the method comprising the following steps:

[0007] The pressure and rotation speed data of the underwater dredging robot are acquired in real time on the left and right sides respectively.

[0008] The changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point are analyzed. The trend values ​​of pressure difference and rotational speed difference at each time point are determined to determine the first trend value of the underwater dredging robot at each time point. Based on the correlation of the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point, the second trend value of the underwater dredging robot at each time point is determined. Combined with the first trend value, the first gain adjustment coefficient of the underwater dredging robot at each time point is determined.

[0009] By measuring the positional distance of the underwater dredging robot at each moment and all moments within the previous preset time period, the trajectory deviation distance of the underwater dredging robot at each moment is determined; based on the changing trend of the trajectory deviation distance at all moments within the previous preset time period, the second gain adjustment coefficient of the underwater dredging robot at the current moment is determined, and combined with the first gain adjustment coefficient, the gain adjustment coefficient of the underwater dredging robot at the current moment is determined.

[0010] Based on the aforementioned gain adjustment coefficient, the gain coefficient of the sliding mode controller inside the underwater dredging robot is optimized to control the attitude of the underwater dredging robot within a preset time period after the current moment.

[0011] Preferably, determining the pressure difference trend value and the rotational speed difference trend value at each time point includes:

[0012] The pressure data differences and rotation speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point are used as inputs to the time series segmentation algorithm, and the output is the trend term sequence of the pressure data difference and the trend term sequence of the rotation speed data difference.

[0013] Two fitting curves are obtained by fitting all elements in the two trend term sequences respectively. The absolute values ​​of the slopes of the fitting curves corresponding to the pressure data differences and the absolute values ​​of the slopes of the fitting curves corresponding to the speed data differences are used as the pressure difference trend values ​​and speed difference trend values ​​at each time point.

[0014] Preferably, the first trend value of the underwater dredging robot at each time point is the result of a positive fusion of the pressure difference trend value and the rotation speed difference trend value at each time point.

[0015] Preferably, the second trend value of the underwater dredging robot at each time point is the inverse of the correlation coefficient of the pressure data difference and rotation speed data difference between the left and right sides of the underwater dredging robot at all times within a preset time period prior to each time point in terms of their changing trends.

[0016] Preferably, the first gain adjustment coefficient of the underwater dredging robot at each time point is the result of positively fusing the normalized value of the first trend value and the second trend value of the underwater dredging robot at each time point.

[0017] Preferably, the method for determining the trajectory deviation distance of the underwater dredging robot at each time point is as follows:

[0018] Calculate the distance between the position of the underwater dredging robot at each time point and the position of the underwater dredging robot at all times within the previous preset time period. Connect the positions of the underwater dredging robots corresponding to the first two distances in the results of the distance sorted in ascending order. The shortest distance from the position of the underwater dredging robot at each time point to the connecting line is taken as the track deviation distance of the underwater dredging robot at each time point.

[0019] Preferably, the method for determining the second gain adjustment coefficient of the underwater dredging robot at the current moment is as follows:

[0020] The trajectory deviation distances of the underwater dredging robot at all times within the preset time period before the current time are used as input to the moving standard deviation algorithm. The output is a moving standard deviation sequence. The fitted straight line obtained by fitting all elements in the moving standard deviation sequence is denoted as the trajectory deviation line. The normalized value of the slope of the trajectory deviation line is used as the second gain adjustment coefficient of the underwater dredging robot at the current time.

[0021] Preferably, the gain adjustment coefficient of the underwater dredging robot at the current moment is the average of the first gain adjustment coefficient and the second gain adjustment coefficient of the underwater dredging robot at the current moment.

[0022] Preferably, the gain coefficient of the optimized sliding mode controller in the underwater dredging robot includes:

[0023] Gain coefficient optimization value within a preset time period after the current time. The expression is: In the formula, This represents the gain adjustment coefficient of the underwater dredging robot at the current moment; Indicates the preset additive factor; The function indicates the preset multiplicative factor; round() indicates the rounding function.

[0024] Preferably, controlling the attitude of the underwater dredging robot within a preset time period after the current moment includes:

[0025] The optimized gain coefficient value for a preset time period after the current moment is used as the gain coefficient of the sliding mode controller inside the underwater dredging robot to control the attitude of the underwater dredging robot for the preset time period after the current moment.

[0026] This application has at least the following beneficial effects:

[0027] This application analyzes the changing trends of the drive wheel speeds of the tracks on both sides of an underwater dredging robot and the external resistance on both sides of its front-end articulated suction device over time as it moves along the dredging cruise path, constructing first and second trend values ​​respectively. Based on the obtained first and second trend values, the application evaluates the hysteresis heading error caused by internal and external disturbances during the underwater dredging robot's movement along the dredging cruise path. This allows for the construction of a first gain adjustment coefficient to adjust the gain coefficient of the sliding mode controller used in the subsequent cruise attitude control module. This improves the response speed of the underwater dredging robot's cruise attitude control module to the hysteresis heading error caused by internal and external disturbances, thereby enabling timely response to underwater... The course of the dredging robot is corrected to avoid deviating from its dredging target. Furthermore, this application constructs a second gain adjustment coefficient by analyzing the distance deviation between the actual navigation trajectory of the underwater dredging robot and its ideal dredging cruise path. This coefficient is used to adjust the gain coefficient of the sliding mode controller used in the subsequent cruise attitude of the underwater dredging robot. This improves the response speed of the subsequent cruise attitude control module of the underwater dredging robot to the hysteresis heading error caused by internal and external disturbances, while also improving the stability of the sliding mode controller. This reduces the instability of the underwater dredging robot due to large heading jitter when moving along the dredging cruise path, thereby improving the dredging efficiency of the underwater dredging robot in complex environments. Attached Figure Description

[0028] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 A flowchart illustrating the steps of an intelligent cruise control method for an underwater dredging robot based on environmental perception, provided in one embodiment of this application.

[0030] Figure 2 This is a schematic diagram illustrating the gain adjustment coefficient extraction process provided in one embodiment of this application. Detailed Implementation

[0031] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the intelligent cruise control method for underwater dredging robots based on environmental perception proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0033] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent cruise control method for underwater dredging robots based on environmental perception provided in this application.

[0034] This application provides an embodiment of an intelligent cruise control method for an underwater dredging robot based on environmental perception. Specifically, the method is described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0035] Step S1: Acquire pressure and rotation speed data on the left and right sides of the underwater dredging robot in real time.

[0036] In this embodiment, the underwater dredging robot is a tracked robot with a suction device. The suction device is used to break, stir and suck up silt and debris from the bottom of the water. A flexible mudguard is installed at the bottom rear end of the suction cover of the suction device to prevent silt and debris from flowing out from the bottom of the suction cover and for subsequent pressure monitoring. The movement speed of the tracks on both sides of the underwater dredging robot is controlled by the drive wheels inside the tracks.

[0037] A sliding plate is installed on each side of the bottom of the underwater dredging robot's suction shroud, connected by springs. A pressure sensor is installed on the spring of each sliding plate to monitor the pressure data on the left and right sides of the robot's front end during underwater movement, reflecting changes in resistance from silt or water flow. A speed sensor is installed on the drive wheels within the tracks on both sides of the underwater dredging robot to monitor the speed data of the tracks during underwater movement. The robot's GPS sensor is used to monitor its position coordinates in real time. The acquisition frequency of all the above data is f. In this embodiment, f is set to 10Hz. In practical applications, as other implementation methods, the implementer can set the frequency according to specific circumstances; this embodiment does not impose any special limitations.

[0038] Furthermore, to eliminate the influence of data dimensions on subsequent analysis and calculations, this embodiment normalizes the pressure and rotational speed data respectively. Many common normalization methods exist; this embodiment uses the maximum-minimum normalization method. In practical applications, as other implementation methods, implementers may also use other normalization methods such as z-score standardization, depending on the specific circumstances. This embodiment does not impose any special restrictions on the selection of normalization methods. The maximum-minimum normalization method is a well-known technique, and the specific process of using it to normalize data will not be elaborated further.

[0039] It should be noted that, unless otherwise specified, all normalization processes in this embodiment employ the maximum-minimum value normalization method.

[0040] Step S2: Analyze the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within the preset time period before each time point, and determine the pressure difference trend value and rotational speed difference trend value at each time point to determine the first trend value of the underwater dredging robot at each time point; based on the correlation of the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within the preset time period before each time point, determine the second trend value of the underwater dredging robot at each time point, and combine it with the first trend value to determine the first gain adjustment coefficient of the underwater dredging robot at each time point.

[0041] Tracked underwater dredging robots employ dual-wheel differential steering, with their heading determined by the speed difference between the left and right tracks. However, the actual propulsion capability of these robots is not only controlled by speed but also significantly affected by external resistance, such as silt and water flow, encountered by the front-end chuck. When the speed difference and pressure difference between the two sides of the robot exhibit opposite trends—for example, the left track has a higher speed than the right track and this speed is increasing—the robot's heading will veer to the right. Simultaneously, the speed of the left track is also affected by resistance from silt and water flow in the robot's forward direction, causing the speed to decrease as resistance increases. Therefore, when the pressure difference and speed difference between the two tracks exhibit opposite trends, the actual propulsion capability of the left track tends to be stronger than that of the right track, further exacerbating the rightward veer of the robot's heading. In other words, when the pressure difference and speed difference exhibit opposite trends, the heading deviation can be far greater than the difference, resulting in control lag and untimely correction.

[0042] Therefore, in order to improve the response speed of the aforementioned hysteretic heading errors caused by internal and external disturbances and achieve rapid and accurate heading correction, this embodiment analyzes the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point. The pressure difference trend value and rotational speed difference trend value at each time point are determined to establish the first trend value of the underwater dredging robot at each time point. Based on the correlation between the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point, the second trend value of the underwater dredging robot at each time point is determined. Combined with the first trend value, the first gain adjustment coefficient of the underwater dredging robot at each time point is determined, laying the foundation for subsequent attitude control of the underwater dredging robot. The specific process is as follows:

[0043] First, this embodiment analyzes the changing trends of pressure data differences between the left and right sides of the underwater dredging robot and the changing trends of rotational speed data differences at all times within a preset time period prior to each time point, and determines the pressure difference trend value and rotational speed difference trend value at each time point. Specifically:

[0044] In this embodiment, the pressure data differences and rotation speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time are used as inputs to the time series segmentation algorithm, and the output is a trend sequence of pressure data differences and a trend sequence of rotation speed data differences.

[0045] Two fitting curves are obtained by fitting all elements in the two trend term sequences respectively. The absolute values ​​of the slopes of the fitting curves corresponding to the pressure data differences and the absolute values ​​of the slopes of the fitting curves corresponding to the speed data differences are used as the pressure difference trend values ​​and speed difference trend values ​​at each time point.

[0046] It should be understood that the preset duration is set manually. In this embodiment, the preset duration is 5 seconds. In actual applications, as other implementation methods, implementers can also set it according to specific circumstances. This embodiment does not impose any special restrictions.

[0047] It should be noted that there are many commonly used time series decomposition algorithms. In this embodiment, the STL time series decomposition algorithm is used to obtain the trend term sequence of pressure data difference and speed data difference. In actual application, as a specific implementation method, implementers may also use other algorithms such as the SEATS decomposition method according to specific circumstances. This embodiment does not impose any special restrictions on the selection of time series decomposition algorithms.

[0048] It should be noted that in this embodiment, the pressure data difference between the left and right sides can be the difference between the pressure data between the left and right sides, or the difference between the pressure data between the right and left sides. If the difference between the pressure data between the left and right sides is calculated as the pressure data difference between the left and right sides, then in subsequent calculations of various differences between the left and right sides, the difference between the left and right sides will be calculated. In order to maintain consistency, this embodiment uniformly uses the result of subtracting the pressure data of the right side from the pressure data of the left side as the pressure data difference between the left and right sides, and uses the result of subtracting the speed data of the right side from the speed data of the left side as the speed data difference between the left and right sides.

[0049] It should be noted that there are many commonly used fitting methods. In this embodiment, the least squares method is used to fit the differences in pressure data and speed data respectively. In practical applications, as other implementation methods, implementers may also use other fitting algorithms such as polynomial function fitting method according to specific circumstances. This embodiment does not impose any special restrictions on the selection of fitting algorithms.

[0050] The STL time series decomposition algorithm and the least squares method are both well-known techniques. The specific process of obtaining the trend term sequence using the STL time series decomposition algorithm and the specific process of fitting the pressure data difference and speed data difference using the least squares method will not be described in detail.

[0051] Furthermore, this embodiment determines the first trend value of the underwater dredging robot at each time point based on the pressure difference trend value and the rotation speed difference trend value at each time point, specifically:

[0052] In this embodiment, the result of positively fusing the pressure difference trend value and the rotation speed difference trend value at each time point is used as the first trend value of the underwater dredging robot at each time point.

[0053] It should be understood that positive fusion refers to combining two or more indicators through addition or multiplication to obtain a comprehensive indicator, thereby more comprehensively and accurately assessing a phenomenon or problem. This fusion method is not limited to simple arithmetic operations, but can also include more complex statistical models and analytical methods. Implementers can choose according to specific circumstances, and this embodiment does not impose any special restrictions.

[0054] Preferably, as one implementation method, the average of the pressure difference trend value and the rotation speed difference trend value at each time point is used as the first trend value of the underwater dredging robot at each time point. In actual application, as another implementation method, the implementer may also adopt other positive fusion methods such as sum or product according to the specific situation. This embodiment does not impose any special restrictions.

[0055] Based on the first trend value of the underwater dredging robot at each time point, it can be understood that the first trend value is a comprehensive quantitative assessment of the degree of change in the robot's internal driving factors, namely the difference in rotational speed between the two tracks, and external environmental resistance factors, namely the difference in pressure between the two sides of the robot's front-end articulator. The first trend characteristic value reflects the overall trend strength of the two key variables, rotational speed difference and pressure difference, over time, and is used to assess how strong the potential driving force is for the underwater dredging robot to face the risk of heading deviation. If the absolute value of the slope of the fitted curve corresponding to the pressure data difference within a preset time period before the current time is larger, and the absolute value of the slope of the fitted curve corresponding to the rotational speed data difference is larger, it means that both the pressure difference and the rotational speed difference have shown a unidirectional, continuous, and significant increasing or decreasing trend in the period before the current time. The larger the corresponding trend characteristic value, the greater the value. This reflects that the robot's internal driving system or external force is undergoing drastic changes, indicating that a steadily increasing heading disturbance torque is forming. If not intervened in time, it will cause the underwater dredging robot to deviate from its heading seriously and continuously.

[0056] Conversely, if the absolute value of the slope of the fitting curve corresponding to the pressure data difference within the preset time period before the current moment is smaller, and the absolute value of the slope of the fitting curve corresponding to the rotation speed data difference is smaller, this reflects that the internal and external environment of the underwater robot is relatively stable, the interference distance is random or short-lived, the risk of heading deviation is low, and the sliding mode controller in the underwater dredging robot does not need to make a drastic response.

[0057] Furthermore, this embodiment determines the second trend value of the underwater dredging robot at each moment based on the correlation between the pressure data differences and rotation speed data differences between the left and right sides of the underwater dredging robot at all moments within a preset time period prior to each moment. Specifically:

[0058] In this embodiment, the inverse of the correlation coefficient between the pressure data difference and the rotation speed data difference between the left and right sides of the underwater dredging robot at all times within a preset time period before each time is used as the second trend value of the underwater dredging robot at each time.

[0059] In addition, this embodiment uses the inverse of the correlation coefficient between the trend sequence of pressure data differences between the left and right sides of the underwater dredging robot and the trend sequence of rotation speed data differences at all times within a preset time period before each time as the inverse of the correlation coefficient between the pressure data differences and rotation speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time, which is the second trend value of the underwater dredging robot at each time.

[0060] It should be noted that there are many commonly used correlation coefficient calculation methods. In this embodiment, the Pearson correlation coefficient between the trend sequence of pressure data differences and the trend sequence of rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point is used as the correlation coefficient between the trend sequence of pressure data differences and the trend sequence of rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point. In practical applications, as other implementation methods, implementers may also use Spearman correlation coefficient or Kendall rank correlation coefficient, etc., depending on the specific circumstances. This embodiment does not impose any special restrictions on the selection of correlation coefficient calculation methods.

[0061] The calculation method for the Pearson correlation coefficient is a well-known technique, and its specific calculation process will not be elaborated here.

[0062] Specifically, if the correlation coefficient between the trend sequence of pressure data differences between the left and right sides of the underwater dredging robot and the trend sequence of rotational speed data differences is greater than or equal to 0 at all times within the preset time period before each time, then the first trend value is assigned to 0, indicating that the pressure data differences and rotational speed data differences do not have opposite trends.

[0063] Based on the second trend value of the underwater dredging robot at various time points, it can be understood that the second trend value is an assessment of the direction of the coupling effect between the internal driving disturbance and the external environmental resistance disturbance of the underwater dredging robot. The second trend value reflects whether the two key variables, pressure difference and rotational speed difference, show a mutually reinforcing relationship in their changing trends. It is used to assess whether the internal and external factors causing the course deviation are weak in the same direction, thereby exacerbating the risk of course deviation. The magnitude of the second trend value is determined by the negative number of the correlation coefficient between the trend term sequence of pressure data difference and the trend term sequence of rotational speed data difference. The larger the negative number of the correlation coefficient, the smaller and more negative the original correlation coefficient is, that is, the stronger the negative correlation between pressure data difference and rotational speed data difference, and the larger the corresponding second trend value. This reflects a very dangerous situation: for example, when the rotational speed of the left track continuously increases relative to the right, the resistance of the left traction device continuously decreases relative to the right. The superposition of these two trends will drastically amplify the net propulsion force of the left track, causing the underwater dredging robot's tendency to deviate to the right to be drastically amplified. The internal and external disturbances resonate, and the risk of course deviation is extremely high.

[0064] Conversely, if the negative number of the correlation coefficient is smaller or close to 0, it means that the difference in pressure data and the difference in speed data are positively correlated or uncorrelated in terms of their changing trends. The smaller the corresponding second trend value, the more it reflects that the internal and external disturbances are independent or cancel each other out. They will not form a combined force to exacerbate the course deviation, and may even neutralize each other. Therefore, the coupling risk of course deviation is low.

[0065] Furthermore, this embodiment determines the first gain adjustment coefficient of the underwater dredging robot at each time point based on the first trend value and the second trend value, specifically:

[0066] In this embodiment, the normalized value of the first trend value of the underwater dredging robot at each time point and the result of positively fusing the second trend value are used as the first gain adjustment coefficient of the underwater dredging robot at each time point.

[0067] Preferably, as an implementation method, in this embodiment, the product of the normalized value of the first trend value of the underwater dredging robot at each time point and the second trend value is used as the first gain adjustment coefficient of the underwater dredging robot at each time point.

[0068] Based on the first gain adjustment coefficient of the underwater dredging robot at each moment, it can be understood that the first gain adjustment coefficient reflects the degree and danger of the deviation risk. It is used to assess how much the gain coefficient of the sliding mode controller needs to be increased in the next control moment or cycle in order to quickly suppress the impending hysteresis heading error. If the first trend value of the underwater dredging robot at the current moment is larger, the first gain adjustment coefficient will be larger. This reflects that the changing trend of internal and external disturbances of the robot is very drastic. Therefore, the sliding mode controller needs to intervene earlier with a faster response speed, that is, a larger gain, to strongly correct the deviation and avoid a large hysteresis deviation in heading. At the same time, if the second trend value at the current moment is larger, the first gain adjustment coefficient will also increase. This reflects that internal and external disturbances are forming a combined force, exacerbating the heading deviation risk of the underwater dredging robot. This is a high-risk signal, requiring the controller to intervene with the greatest vigilance and the fastest speed to prevent the heading from rapidly getting out of control.

[0069] Conversely, if the first trend value of the underwater dredging robot is smaller at the current moment, the first gain adjustment coefficient is smaller. This reflects that the changing trend of internal and external disturbances of the robot tends to be gentle, and the risk of a significant lag deviation in the future course is low. Therefore, the sliding mode controller does not need to intervene too early or too strongly and can maintain a small gain, thereby avoiding unnecessary energy consumption and system jitter caused by over-control. At the same time, if the second trend value is smaller at the current moment, the first gain adjustment coefficient also decreases. This reflects that internal and external disturbances not only do not form a combined force, but may even cancel each other out, weakening the overall impact on the course. This indicates that the current navigation state of the underwater dredging robot is stable, and the controller can maintain a "wait-and-see" attitude without increasing the gain, ensuring the smoothness and efficiency of the cruise process.

[0070] Thus, this embodiment constructs a first gain adjustment coefficient that can dynamically assess the coupling risk of internal and external disturbances by comprehensively analyzing the trend intensity and correlation of the difference between pressure and rotation speed. This enables early prediction and rapid suppression of hysteresis heading errors, which helps to improve the response speed and dredging efficiency of underwater dredging robots in complex and variable environments.

[0071] Step S3: By measuring the positional distance of the underwater dredging robot at each moment and all moments within the previous preset time period, determine the trajectory deviation distance of the underwater dredging robot at each moment; based on the changing trend of the trajectory deviation distance at all moments within the previous preset time period, determine the second gain adjustment coefficient of the underwater dredging robot at the current moment, and combine it with the first gain adjustment coefficient to determine the gain adjustment coefficient of the underwater dredging robot at the current moment.

[0072] The gain coefficient in the sliding mode controller of the underwater dredging robot is a key parameter affecting the dynamic performance of the sliding mode controller: the larger the gain, the faster the error convergence speed of the sliding mode controller, but excessive gain will cause overshoot and chattering, which will destroy the cruise stability of the underwater dredging robot. Therefore, this embodiment adaptively adjusts the gain coefficient. That is, when the deviation distance between the actual route and the ideal path of the underwater dredging robot is detected to fluctuate significantly within the current sampling period, it indicates that the current gain coefficient is too high and has caused the sliding mode controller to show an unstable trend. In order to suppress chattering and improve the stability of the sliding mode controller, the gain coefficient is automatically reduced, thereby ensuring the smooth cruise attitude of the underwater dredging robot.

[0073] Therefore, based on the above analysis, this embodiment determines the trajectory deviation distance of the underwater dredging robot at each moment by measuring the positional distance between each moment and all moments within a preset time period. Based on the changing trend of the trajectory deviation distance at all moments within a preset time period before the current moment, the second gain adjustment coefficient of the underwater dredging robot at the current moment is determined, and combined with the first gain adjustment coefficient, the gain adjustment coefficient of the underwater dredging robot at the current moment is determined, so as to adaptively adjust the gain coefficient of the sliding mode controller. The specific process is as follows:

[0074] In this embodiment, firstly, by measuring the positional distance of the underwater dredging robot at each moment and all moments within a preset time period, the trajectory deviation distance of the underwater dredging robot at each moment is determined, specifically as follows:

[0075] This embodiment calculates the distance between the position of the underwater dredging robot at each time point and all times within a preset time period. The distances are sorted in ascending order, and the positions of the underwater dredging robots corresponding to the first two distances are connected. The shortest distance from the position of the underwater dredging robot at each time point to the connected line is taken as the track deviation distance of the underwater dredging robot at each time point.

[0076] Based on the trajectory deviation distance of the underwater dredging robot at each moment, it can be understood that the trajectory deviation distance reflects the trajectory tracking accuracy of the underwater dredging robot at each moment, and is used to characterize the effectiveness of the underwater dredging robot's cruise attitude control, that is, whether the underwater dredging robot is accurately traveling on the predetermined route. If the shortest distance from the position of the underwater dredging robot to the line at the current moment is larger, it indicates that the actual position of the underwater dredging robot deviates from the ideal path, and the corresponding trajectory deviation distance is larger. This reflects that the actual course of the underwater dredging robot is far from the predetermined task path, indicating that there is a large error in the attitude control of the underwater dredging robot at the current moment.

[0077] Conversely, the smaller the shortest distance from the current position of the underwater dredging robot to the connecting line, the closer the actual position of the underwater dredging robot is to the ideal path, and the smaller the corresponding trajectory deviation distance. This reflects that the actual heading of the underwater dredging robot is highly consistent with the predetermined mission path, indicating that the attitude control of the underwater dredging robot at the current moment is precise and effective, and it can complete the cruise tracking mission well.

[0078] Furthermore, this embodiment determines the second gain adjustment coefficient of the underwater dredging robot at the current moment based on the changing trend of the track deviation distance at all times within a preset time period prior to the current moment, specifically as follows:

[0079] In this embodiment, the trajectory deviation distance of the underwater dredging robot at all times within a preset time period before the current time is used as the input of the moving standard deviation algorithm. The length of the sliding window in the moving standard deviation algorithm is set to L. In this embodiment, L is set to 1s. In actual application, the implementer can also set it according to the specific situation. Finally, the moving standard deviation sequence is output. The fitted straight line obtained by fitting all elements in the moving standard deviation sequence is denoted as the trajectory deviation line. The normalized value of the slope of the trajectory deviation line is used as the second gain adjustment coefficient of the underwater dredging robot at the current time.

[0080] It should be noted that the least squares fitting method is used to fit all elements in the moving standard deviation sequence. In practical applications, as other implementation methods, implementers may also use other fitting methods according to specific circumstances. This embodiment does not impose any special restrictions.

[0081] Specifically, if the slope of the trajectory deviation line is 0, then the second gain adjustment coefficient of the corresponding underwater dredging robot is assigned a value of 0.

[0082] It should be noted that the preset time period length is set manually. In this embodiment, the preset time period length is 5 seconds. In actual applications, as other implementation methods, implementers can also set it according to specific circumstances. This embodiment does not impose any special restrictions.

[0083] The moving standard deviation algorithm is a well-known technique, and the specific process of obtaining the moving standard deviation sequence using it will not be described in detail.

[0084] Based on the second gain adjustment coefficient of the underwater dredging robot at the current moment, it can be understood that the second gain adjustment coefficient reflects the direction of change of the fluctuation of the track deviation distance over time. It is used to characterize or evaluate whether the gain setting of the current sliding mode controller is too large, thereby causing instability or chattering of the underwater dredging robot. If the normalized value of the slope of the track deviation line at the current moment is larger, it means that the fluctuation of the track deviation has shown a continuous and significant increasing trend in the period before the current moment. The larger the track deviation characteristic value, this reflects that although the underwater dredging robot may be close to the path, its course swings back and forth and oscillates violently near the path. This is a typical overshoot and chattering performance caused by excessive gain. The gain must be reduced to increase damping and make the navigation of the underwater dredging robot smoother.

[0085] Conversely, the smaller the normalized value of the slope of the trajectory deviation line at the current moment, the more stable or decreasing the fluctuation of the trajectory deviation has been in the period before the current moment. The smaller the trajectory deviation characteristic value, the more it reflects that the underwater dredging robot's navigation trajectory converges smoothly and stably to or remains near the predetermined path without obvious oscillation or overshoot. This indicates that the current gain setting of the sliding mode controller is appropriate, and the system has good stability and damping characteristics. There is no need to reduce the gain to suppress chattering, and it is even possible to consider appropriately increasing the gain to enhance the response speed when necessary.

[0086] Furthermore, this embodiment determines the gain adjustment coefficient of the underwater dredging robot at each time point based on the first gain adjustment coefficient and the second gain adjustment coefficient, specifically:

[0087] In this embodiment, the average of the first gain adjustment coefficient and the second gain adjustment coefficient of the underwater dredging robot at the current moment is used as the gain adjustment coefficient of the underwater dredging robot at the current moment.

[0088] Preferably, the schematic diagram of the gain adjustment coefficient extraction process provided in this embodiment is as follows: Figure 2 As shown.

[0089] Based on the gain adjustment coefficient of the underwater dredging robot at the current moment, it can be understood that the gain adjustment coefficient reflects the ideal adjustment weight of the sliding mode controller gain coefficient in the subsequent control process. It is used to characterize how to achieve a balance between the two contradictory requirements of "suppressing the impending course deviation" and "avoiding self-induced underwater robot jitter". If the first gain adjustment coefficient at the current moment is larger, the gain adjustment coefficient will also increase. This reflects that the underwater dredging robot predicts that it will face a severe course deviation challenge in the future, requiring the controller to prioritize speed and increase the gain to improve the response speed, at the cost of some stability to ensure that it does not deviate from the path. At the same time, if the second gain adjustment coefficient at the current moment is larger, the corresponding gain adjustment coefficient should also increase. This indicates that the robot is exhibiting unstable jitter due to excessive gain. This is a strong signal to reduce the gain, requiring the controller to prioritize stability and suppress jitter by reducing the gain.

[0090] Conversely, if the first gain adjustment coefficient is smaller at the current moment, the gain adjustment coefficient will decrease accordingly. This reflects that the underwater dredging robot predicts a low risk of future course deviation and that internal and external disturbances tend to be mild. Therefore, there is no need to overemphasize rapid response. The sliding mode controller can prioritize maintaining the current smaller gain to ensure smooth navigation and energy saving. At the same time, if the second gain adjustment coefficient is smaller at the current moment, the corresponding gain adjustment coefficient should also decrease. This indicates that the current system is operating stably and there are no signs of chattering caused by excessive gain. The sliding mode controller does not need to actively reduce the gain to suppress chattering.

[0091] Thus, this embodiment constructs a gain adjustment coefficient that can intelligently balance control agility and stability by dynamically fusing a first gain adjustment coefficient that reflects the risk of heading deviation and a second gain adjustment coefficient that characterizes the system's chattering trend. This enables adaptive optimization of the sliding mode controller gain, thereby effectively suppressing heading deviation while avoiding chattering caused by excessive gain, and significantly improving the dredging efficiency of the underwater dredging robot in complex environments.

[0092] Step S4: Based on the gain adjustment coefficient, optimize the gain coefficient of the sliding mode controller inside the underwater dredging robot to control the attitude of the underwater dredging robot within a preset time period after the current moment.

[0093] Based on the gain adjustment coefficient obtained in step S3, the gain coefficient of the sliding mode controller inside the underwater dredging robot is optimized to control the attitude of the underwater dredging robot within a preset time period after the current moment. Specifically:

[0094] As one implementation method, in this embodiment, the optimized gain coefficient value within a preset time period after the current time is... The expression is: In the formula, This represents the gain adjustment coefficient of the underwater dredging robot at the current moment; Indicates the preset additive factor; The function indicates the preset multiplicative factor; round() indicates the rounding function.

[0095] It should be noted that the values ​​of the preset additive factor and the preset multiplicative factor are both set manually. In this embodiment, the preset additive factor is 10 and the preset multiplicative factor is 90. In actual applications, as other implementation methods, implementers can also set them according to specific circumstances. This embodiment does not impose any special restrictions.

[0096] Based on the optimized gain coefficient value, it can be understood that if the gain adjustment coefficient is larger at the current moment, it indicates that a stronger correction capability is needed and the current jitter risk is low. Therefore, the optimized gain coefficient value is increased to respond quickly to errors. Conversely, if the gain adjustment coefficient is smaller at the current moment, it indicates that the current correction requirement is weak or the jitter risk is high. Therefore, the optimized gain coefficient value is reduced to prioritize the stability of system operation and avoid unnecessary oscillations or overshoots caused by excessive gain, thereby ensuring that the underwater dredging robot can cruise smoothly and safely along the predetermined path.

[0097] Furthermore, the optimized gain coefficient value within a preset time period after the current moment is used as the gain coefficient of the sliding mode controller inside the underwater dredging robot to control the attitude of the underwater dredging robot within the preset time period after the current moment, wherein the attitude is the yaw angle of the underwater dredging robot.

[0098] The working principle of the sliding mode controller is a well-known technology, and the specific process of using it to control the posture of the underwater dredging robot will not be described in detail.

[0099] Thus, this embodiment achieves precise and dynamic control of the yaw angle of the underwater dredging robot by converting the gain adjustment coefficient into the optimized gain coefficient of the sliding mode controller. This effectively balances the system's requirements for rapid response and stable operation in complex environments, and significantly improves the accuracy and reliability of cruise path tracking.

[0100] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0101] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0102] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An intelligent cruise control method for underwater dredging robots based on environmental perception, characterized in that, The method includes the following steps: The pressure and rotation speed data of the underwater dredging robot are acquired in real time on the left and right sides respectively. The changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point are analyzed. The trend values ​​of pressure difference and rotational speed difference at each time point are determined to determine the first trend value of the underwater dredging robot at each time point. Based on the correlation of the changing trends of pressure data differences and rotational speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point, the second trend value of the underwater dredging robot at each time point is determined. Combined with the first trend value, the first gain adjustment coefficient of the underwater dredging robot at each time point is determined. By measuring the positional distance of the underwater dredging robot at each moment and all moments within the previous preset time period, the trajectory deviation distance of the underwater dredging robot at each moment is determined; based on the changing trend of the trajectory deviation distance at all moments within the previous preset time period, the second gain adjustment coefficient of the underwater dredging robot at the current moment is determined, and combined with the first gain adjustment coefficient, the gain adjustment coefficient of the underwater dredging robot at the current moment is determined. Based on the aforementioned gain adjustment coefficient, the gain coefficient of the sliding mode controller inside the underwater dredging robot is optimized to control the attitude of the underwater dredging robot within a preset time period after the current moment. The method for determining the trajectory deviation distance of the underwater dredging robot at each time point is as follows: Calculate the distance between the position of the underwater dredging robot at each time point and the position of the underwater dredging robot at all times within the previous preset time period. Connect the positions of the underwater dredging robots corresponding to the first two distances in the results of the distance sorted in ascending order. The shortest distance from the position of the underwater dredging robot at each time point to the connecting line is taken as the track deviation distance of the underwater dredging robot at each time point. The method for determining the second gain adjustment coefficient of the underwater dredging robot at the current moment is as follows: The trajectory deviation distances of the underwater dredging robot at all times within the preset time period before the current time are used as input to the moving standard deviation algorithm. The output is a moving standard deviation sequence. The fitted straight line obtained by fitting all elements in the moving standard deviation sequence is denoted as the trajectory deviation line. The normalized value of the slope of the trajectory deviation line is used as the second gain adjustment coefficient of the underwater dredging robot at the current time.

2. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The determination of the pressure difference trend value and the speed difference trend value at each time point includes: The pressure data differences and rotation speed data differences between the left and right sides of the underwater dredging robot at all times within a preset time period before each time point are used as inputs to the time series segmentation algorithm, and the output is the trend term sequence of the pressure data difference and the trend term sequence of the rotation speed data difference. Two fitting curves are obtained by fitting all elements in the two trend term sequences respectively. The absolute values ​​of the slopes of the fitting curves corresponding to the pressure data differences and the absolute values ​​of the slopes of the fitting curves corresponding to the speed data differences are used as the pressure difference trend values ​​and speed difference trend values ​​at each time point.

3. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The first trend value of the underwater dredging robot at each time point is the result of a positive fusion of the pressure difference trend value and the rotation speed difference trend value at each time point.

4. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The second trend value of the underwater dredging robot at each time point is the inverse of the correlation coefficient of the pressure data difference and rotation speed data difference between the left and right sides of the underwater dredging robot at all times within a preset time period prior to each time point.

5. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The first gain adjustment coefficient of the underwater dredging robot at each time point is the result of the positive fusion of the normalized value of the first trend value and the second trend value of the underwater dredging robot at each time point.

6. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The gain adjustment coefficient of the underwater dredging robot at the current moment is the average of the first gain adjustment coefficient and the second gain adjustment coefficient of the underwater dredging robot at the current moment.

7. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The gain coefficient of the optimized sliding mode controller in the underwater dredging robot includes: Gain coefficient optimization value within a preset time period after the current time. The expression is: In the formula, This represents the gain adjustment coefficient of the underwater dredging robot at the current moment; Indicates the preset additive factor; The function indicates the preset multiplicative factor; round() indicates the rounding function.

8. The intelligent cruise control method for underwater dredging robots based on environmental perception as described in claim 1, characterized in that, The control of the underwater dredging robot's attitude over a preset time period following the current moment includes: The optimized gain coefficient value for a preset time period after the current moment is used as the gain coefficient of the sliding mode controller inside the underwater dredging robot to control the attitude of the underwater dredging robot for the preset time period after the current moment.

Images