Self-adaptive control method and system for underwater operation power of dredging robot

Abstract

The invention relates to the field of robot power self-adaptive control, in particular to a dredging robot underwater operation power self-adaptive control method and system, and aims to fully consider underwater operation complexity when the dredging robot underwater operation power self-adaptive control method and system are realized. According to elevation data, a to-be-operated area is graded according to operation difficulty, different power adaptive control methods are adopted for different operation difficulty levels, and the problem of difficulty in power adaptive control of nonlinear and multivariable coupled underwater operation scenes is solved by utilizing the advantages of different algorithms for processing different complex scenes.

Description

Technical Field

[0001] The present invention relates to the field of robot power adaptive control, and in particular to a method and system for adaptively controlling the underwater operation power of a dredging robot. Background Art

[0002] In the urban water system, river dredging is a key link in maintaining the health of the water ecology. In recent years, as people have higher and higher requirements for water environment quality, dredging has become a "compulsory course" for large and small rivers. Too much silt deposition will not only affect the water quality, but also when the weather warms up slightly, the bottom mud will rise to the surface of the water, greatly reducing the appearance of the river. Dredging robots are widely used in the field of underwater sediment cleaning. Their core function is to complete the excavation, transportation and treatment of silt through actuators such as robotic arms and suction devices.

[0003] The power control of traditional dredging robots is relatively simple, and the power distribution modes include: fixed power mode, which sets constant power according to preset working conditions, and cannot adapt to dynamically changing environmental parameters such as silt density and water flow velocity, which can easily lead to energy waste or low dredging efficiency. Manual adjustment mode, which relies on the operator's experience to adjust the power, has a delayed response and insufficient accuracy, and is difficult to cope with complex underwater environments. The power feedback adjustment mode is generally a simple feedback control, that is, local feedback adjustment based on a single sensor (such as motor current), lacks multi-parameter collaborative analysis capabilities, and is prone to control failure due to environmental interference. Traditional power control methods cannot adapt to the changing underwater environment, such as uneven silt distribution and undulating terrain. Fixed power or extensive adjustment can easily cause motor overload or inefficient operation, especially under light load conditions, resulting in unnecessary energy loss. Existing control systems mostly rely on linear algorithms, which are difficult to handle nonlinear and multi-variable coupled underwater operation scenarios, and the dynamic response speed and decision-making accuracy are limited. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method and system for adaptively controlling the underwater operation power of a dredging robot, which are used to solve the problems existing in the prior art.

[0005] The present invention provides a method for adaptively controlling the underwater operation power of a dredging robot, comprising: S1: Collect underwater operation data of the dredging robot, including historical operation time and corresponding historical operation area, and determine the relationship between the historical operation time and the historical operation area; S2: Determine the waiting time for operation of the current waiting area according to the relationship between the historical operation time and the area of ​​the operation area, and the area of ​​the waiting area; S3: Scan the terrain of the underwater operation area and estimate the difficulty of the operation; S4: establishing a power allocation model for the dredging robot according to the waiting operation time, the area of ​​the waiting operation area and the operation difficulty, and performing power adaptive control according to the allocation model.

[0006] Preferably, the dredging robot includes a control module, a propulsion device, a dredging device, a conveying device, a communication device, an auxiliary device and a safety device; the auxiliary device includes a camera, a lighting lamp, a sonar, and a lidar; the safety device includes a floating system and a fault protection system.

[0007] Preferably, S1 also includes: the underwater operation data of the dredging robot also includes historical operation power allocation data, and a neural network power adaptive control model is established based on the historical operation time, the historical operation area, and the historical operation power allocation data.

[0008] Preferably, S3 further includes: S31: Divide the area to be operated into N blocks on average, where N>4; use the laser radar to measure the underwater terrain of the area to be operated to obtain N elevation data sets; S32: Collect the elevation data of the lowest point in the N elevation data sets and calculate the average value to obtain the elevation calibration value T of the area to be operated; S33: The control device of the dredging robot is embedded with GIS software, which performs three-dimensional modeling of the bottom silt based on the elevation calibration value T, and calculates the volume Vn of the N blocks of the area to be operated; S34: Classifying the difficulty of the operation of the N areas to be operated into easy, medium and difficult according to the volume Vn.

[0009] Preferably, S4 establishing a power distribution model of the dredging robot includes: S41: the control module obtains the operation difficulty of the area to be operated; S42: the control module allocates power according to the difficulty of the operation; S43: The control module allocates fixed power values ​​to the auxiliary device and the safety device.

[0010] Preferably, the control module in S42 allocates power according to the difficulty of the operation, including: S421: When the operation difficulty is easy, based on the waiting operation time and the waiting operation area, a power adaptive control strategy is determined using a neural network power adaptive control model; S422: When the operation difficulty is medium, the expert PID control model is used to map the waiting operation time, the area of ​​the waiting operation area and the volume Vn into power adjustment parameters, and the power adaptive control strategy is dynamically adjusted; S423: When the operation difficulty is difficult, a reinforcement learning model is used to take the waiting time for operation, the area of ​​the waiting area for operation and the volume Vn as state observation inputs to obtain an optimal power adaptive control strategy to balance performance and energy consumption.

[0011] Preferably, the expert PID control model utilizes domain expert experience or machine learning of underwater operation data of the dredging robot to generate power adaptive control rules.

[0012] Preferably, the mapping is a power adjustment parameter, and the expert PID control model parameters are adjusted by changing the waiting time for operation, the area of ​​the waiting area for operation and the volume Vn.

[0013] Preferably, the reinforcement learning model includes: A state observation module collects the waiting time, the area of ​​the waiting area and the volume Vn to describe the current environment state; An action selection module, based on the current environmental state, determines a current power adaptive control strategy through a predetermined strategy or a random strategy; A reward feedback module, which obtains a reward signal according to the feedback of the current environment state; The strategy update module establishes a deep neural network according to the reward signal to update the parameters of the reinforcement learning and obtain the optimal power adaptive control strategy.

[0014] According to another aspect of the present invention, a dredging robot underwater operation power adaptive control system is provided, the system adopts the above-mentioned dredging robot underwater operation power adaptive control method, the system comprises: A data collection module is used to collect underwater operation data of the dredging robot, including historical operation time and corresponding operation area area, and determine the relationship between the historical operation time and the operation area area; A data processing module is used to process the data collected by the data collection module to obtain a power adaptive control solution; The data processing of the data collected by the data acquisition module includes: Determine the waiting time for the current area to be operated according to the relationship between the historical operation time and the area of ​​the operation area, as well as the area of ​​the area to be operated; Scan the terrain of the underwater operation area and estimate the difficulty of the operation; A power allocation model of the dredging robot is established according to the waiting time for operation, the area of ​​the waiting operation area and the difficulty of the operation, and power adaptive control is performed according to the allocation model.

[0015] The present invention has the following technical effects: When using a dredging robot for underwater operations, the present invention first obtains historical data and processes the data; uses a laser radar to obtain silt elevation data in the river channel and calculates the silt volume in the area to be operated, and classifies the area to be operated according to the difficulty of the operation. Different power adaptive control methods are used for different levels of difficulty of the operation. It can handle nonlinear and multivariable coupled underwater operation scenarios and improve the operating reliability of the dredging robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a flow chart of a method for adaptively controlling the underwater operation power of a dredging robot provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0019] Embodiment 1, attached Figure 1 A flowchart of a method for adaptively controlling the underwater power of a dredging robot is shown in the attached figure. Figure 1 As shown, a method for adaptively controlling the underwater operation power of a dredging robot comprises: S1: Collect underwater operation data of the dredging robot, including historical operation time and corresponding historical operation area, and determine the relationship between the historical operation time and the operation area; In this embodiment, the relationship formula between the historical operation time and the historical operation area can be obtained by linear fitting; The underwater operation data of the dredging robot also includes historical operation power allocation data, and a neural network power adaptive control model is established based on the historical operation time, historical operation area and historical operation power allocation data; in this embodiment, the historical operation time and historical operation area are used as input, and the historical operation power allocation data is used as output to train and establish a neural network power adaptive control model.

[0020] S2: Determine the waiting time for operation of the current waiting area according to the relationship between the historical operation time and the area of ​​the historical operation area, and the area of ​​the waiting area; In this embodiment, the area of ​​the waiting-for-operation area is input into the relationship formula between the historical operation time and the area of ​​the historical operation area, so as to determine the waiting-for-operation time of the current waiting-for-operation area.

[0021] S3: Scan the terrain of the underwater operation area and estimate the difficulty of the operation; The specific steps include: S31: dividing the area to be operated into N blocks on average, where N>4; using a laser radar to measure the underwater terrain of the area to be operated to obtain N elevation data sets; Among them, in this embodiment, the lidar uses a green pulsed laser source to illuminate the target scene, and the reflected pulsed illumination is detected by a single-photon detector array, which can distinguish between photons reflected by the target and photons reflected by particles in the water, making it particularly suitable for 3D imaging in highly turbid water.

[0022] S32: Collect the lowest point elevation data from N elevation data sets and calculate the average value to obtain the elevation calibration value T of the area to be operated; S33: The control module of the dredging robot is embedded with GIS software, which performs three-dimensional modeling of the underwater silt based on the elevation calibration value T, and calculates the silt volume Vn of N blocks of the working area; S34: Classifying the difficulty of the operation in the N areas to be operated into easy, medium and difficult according to the sludge volume Vn.

[0023] In this embodiment, when the silt volume is large during the silt removal operation of the silt removal robot, the silt removal difficulty increases, and the previous power allocation may affect the silt removal efficiency, so the silt volume needs to be considered as a factor for additional power allocation. When the silt volume is large, the power is adaptively adjusted to reduce the power of the underwater propulsion device of the silt removal robot, and allocate more power to the silt cleaning, silt transportation and other devices to complete the silt removal work efficiently and stably.

[0024] S4: establishing a power allocation model of the dredging robot according to the waiting operation time, the area of ​​the waiting operation area and the operation difficulty, and performing power adaptive control according to the allocation model; The specific steps include: S41: the control module obtains the operation difficulty of the area to be operated; S42: the control module allocates power according to the difficulty of the operation; Among them, in this embodiment, when the difficulty of the operation is easy, the power adaptive control strategy is determined by using a neural network control model based on the waiting time and the area of ​​the waiting area; the neural network model is established through training of the historical operation data of the dredging robot. When the difficulty of the operation is easy, the river channel silt is relatively flat, and there is no need to consider the impact of the silt volume on the power adaptive control.

[0025] In this embodiment, when the operation difficulty is medium, the domain expert experience or machine learning dredging robot underwater operation data is used to establish an expert PID control model, and the waiting operation time, the area of ​​the waiting operation area and the volume Vn are mapped to power adjustment parameters to dynamically adjust the power adaptive control strategy. The expert PID control model uses the domain expert experience or machine learning dredging robot underwater operation data to generate power adaptive control rules. The mapping is a power adjustment parameter, and the expert PID control model parameters are adjusted by changing the waiting operation time, the area of ​​the waiting operation area and the volume Vn. The expert PID control model is suitable for nonlinear systems and can handle power distribution under medium operation difficulty.

[0026] In this embodiment, when the operation difficulty is difficult, a reinforcement learning model is used to take the waiting time, the area of ​​the waiting area and the volume Vn as state observation inputs to obtain an optimal power adaptive control strategy to balance performance and energy consumption.

[0027] A state observation module collects the waiting time, the area of ​​the waiting area and the volume Vn to describe the current environment state; An action selection module, based on the current environmental state, determines a current power adaptive control strategy through a predetermined strategy or a random strategy; A reward feedback module, which obtains a reward signal according to the feedback of the current environment state; Exemplarily, when the current environmental state changes, for example, when the area to be operated decreases or the operation time increases, a reward signal is fed back.

[0028] A strategy update module, which establishes a deep neural network according to the reward signal to update the parameters of the reinforcement learning and obtain an optimal power adaptive control strategy; For example, when the area to be operated decreases and the operation time increases, the updated current environmental state is input into a pre-trained deep neural network to output reinforcement learning parameters, and the current power adaptive control strategy can prioritize the allocation of propulsion devices.

[0029] S43: The control module allocates fixed power values ​​to the auxiliary device and the safety device.

[0030] Among them, auxiliary devices include cameras, lighting, sonar, and lidar; safety devices include flotation systems and fault protection systems, which are the key to protecting the safe and stable operation of the dredging robot. Therefore, no matter what the difficulty of the operation, fixed power values ​​are allocated to the auxiliary devices and safety devices.

[0031] Embodiment 2, the present invention further provides a dredging robot underwater operation power adaptive control system, the system adopts a dredging robot underwater operation power adaptive control method of embodiment 1, the system comprises: A data collection module is used to collect underwater operation data of the dredging robot, including historical operation time and corresponding operation area area, and determine the relationship between the historical operation time and the operation area area; A data processing module is used to process the data collected by the data collection module to obtain a power adaptive control solution; The data processing of the data collected by the data acquisition module includes: Determine the waiting time for the current area to be operated according to the relationship between the historical operation time and the area of ​​the operation area, as well as the area of ​​the area to be operated; Scan the terrain of the underwater operation area and estimate the difficulty of the operation; A power allocation model of the dredging robot is established according to the waiting time for operation, the area of ​​the waiting operation area and the difficulty of the operation, and power adaptive control is performed according to the allocation model.

[0032] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0033] It should also be noted that the orientations or positional relationships indicated by the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0034] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for adaptively controlling the underwater operation power of a dredging robot, characterized in that: include: S1: Collect underwater operation data of the dredging robot, including historical operation time and corresponding historical operation area, and determine the relationship between the historical operation time and the operation area; S2: Determine the waiting time for operation of the current waiting area according to the relationship between the historical operation time and the area of ​​the historical operation area, and the area of ​​the waiting area; S3: Scan the terrain of the underwater operation area and estimate the difficulty of the operation; S4: establishing a power allocation model of the dredging robot according to the waiting operation time, the area of ​​the waiting operation area and the difficulty of the operation, and performing power adaptive control according to the allocation model.

2. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 1, characterized in that: The dredging robot includes a control module, a propulsion device, a dredging device, a conveying device, a communication device, an auxiliary device and a safety device; the auxiliary device includes a camera, a lighting lamp, a sonar and a lidar; the safety device includes a floating system and a fault protection system.

3. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 2, characterized in that: The S1 further comprises: The underwater operation data of the dredging robot also includes historical operation power allocation data. Based on the historical operation time, the historical operation area and the historical operation power allocation data, a neural network power adaptive control model is established.

4. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 2, characterized in that: The S3 further includes: S31: Divide the area to be operated into N blocks on average, where N>4; use the laser radar to measure the underwater terrain of the area to be operated in blocks to obtain N elevation data sets; S32: Collect the lowest point elevation data from the N elevation data sets and calculate the average value to obtain the elevation calibration value T of the area to be operated; S33: The control module of the dredging robot is embedded with GIS software, and three-dimensional modeling of the bottom silt is performed based on the elevation calibration value T, and the volume Vn of N blocks of the area to be operated is calculated; S34: Classifying the difficulty of the operation of the N areas to be operated into easy, medium and difficult according to the volume Vn.

5. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 4 is characterized in that: In S4, establishing a power allocation model for the dredging robot includes: S41: the control module obtains the operation difficulty of the area to be operated; S42: the control module allocates power according to the difficulty of the operation; S43: The control module allocates fixed power values ​​to the auxiliary device and the safety device.

6. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 5, characterized in that: In S42, the control module allocates power according to the difficulty of the operation, including: S421: When the operation difficulty is easy, based on the waiting operation time and the waiting operation area, a power adaptive control strategy is determined using a neural network power adaptive control model; S422: When the operation difficulty is medium, the expert PID control model is used to map the waiting operation time, the area of ​​the waiting operation area and the volume Vn into power adjustment parameters, and the power adaptive control strategy is dynamically adjusted; S423: When the operation difficulty is difficult, a reinforcement learning model is used to take the waiting time for operation, the area of ​​the waiting area for operation and the volume Vn as state observation inputs to obtain an optimal power adaptive control strategy to balance performance and energy consumption.

7. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 6, characterized in that: The expert PID control model utilizes domain expert experience or machine learning dredging robot underwater operation data to generate power adaptive control rules.

8. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 7, characterized in that: The mapping is a power adjustment parameter, and the expert PID control model parameters are adjusted by changing the waiting time for operation, the area of ​​the waiting area for operation and the volume Vn.

9. The method for adaptively controlling the underwater operation power of a dredging robot according to claim 6, characterized in that: The reinforcement learning model comprises: A state observation module collects the waiting time, the area of ​​the waiting area and the volume Vn to describe the current environment state; An action selection module, based on the current environmental state, determines a current power adaptive control strategy through a predetermined strategy or a random strategy; A reward feedback module, which obtains a reward signal according to the feedback of the current environment state; The strategy update module establishes a deep neural network according to the reward signal to update the parameters of the reinforcement learning and obtain the optimal power adaptive control strategy.

10. A power adaptive control system for underwater operation of a dredging robot, characterized in that: The system adopts a method for adaptively controlling underwater operation power of a dredging robot according to any one of claims 1 to 9, and the system comprises: A data collection module is used to collect underwater operation data of the dredging robot, including historical operation time and corresponding historical operation area, and determine the relationship between the historical operation time and the historical operation area; A data processing module is used to process the data collected by the data collection module to obtain a power adaptive control solution; The data processing of the data collected by the data acquisition module includes: Determine the waiting time for operation of the current waiting area according to the relationship between the historical operation time and the area of ​​the historical operation area and the area of ​​the waiting area; Scan the terrain of the underwater operation area and estimate the difficulty of the operation; A power allocation model of the dredging robot is established according to the waiting operation time, the area of ​​the waiting operation area and the operation difficulty, and power adaptive control is performed according to the allocation model.

Images