Cooperative operation system and method for photovoltaic cleaning robot and carrying robot

Abstract

The invention discloses a collaborative operation system and method for a photovoltaic cleaning robot and a carrying robot, relates to the technical field of automatic control, and solves the defects that an existing collaborative operation system is unstable in communication and insufficient in emergency handling capacity. According to the invention, the environmental information of the photovoltaic power station is collected in real time through the environmental perception module, and the reliability of data transmission is ensured through an automatic retransmission request mechanism and error detection and correction codes based on 5G and LoRa communication protocols; communication parameters are adaptively adjusted through the communication self-adjusting module, so that the communication quality is ensured; the path planning module dynamically allocates cleaning tasks and plans an optimal path, the collaborative obstacle avoidance module dynamically adjusts the motion trails of the robots and avoids obstacles, the operation progress tracking module records the operation states of the robots in real time, and comprehensive operation progress information is provided through data processing and visual display. According to the invention, the communication stability, the autonomous decision-making capability and the operation efficiency of the system are obviously improved.

Description

Technical Field

[0001] The present invention relates to the field of automation control technology, and more specifically to a collaborative operation system and method of a photovoltaic cleaning robot and a transport robot. Background Art

[0002] With the advancement of global energy transformation and sustainable development, photovoltaic power generation, as an important part of clean energy, has attracted widespread attention for its scale and efficiency. The cleaning of photovoltaic panels is crucial to maintaining power generation efficiency, and the development of automated cleaning technology, especially the collaborative operation of photovoltaic cleaning robots and carrier robots, marks a major change in the maintenance method of photovoltaic power stations. The development of this automated cleaning system not only improves cleaning efficiency, reduces labor costs, but also reduces the impact on the environment. It is one of the key technologies for realizing intelligent management of photovoltaic power stations.

[0003] The document "Design and Research on Automatic Cleaning Robot System for Photovoltaic Modules" mentions a photovoltaic cleaning system, which includes a cleaning trolley, a ferry and a wireless charging system for photovoltaic storage. The cleaning trolley adopts a waterless cleaning method with a spiral roller brush, cleaning while walking. The ferry is responsible for carrying the cleaning trolley and accurately aligning the cleaning trolley with each row of photovoltaic modules through an automatic adjustment mechanism. The wireless charging system for photovoltaic storage is responsible for collecting and storing the power of photovoltaic modules, and charging the batteries of the ferry and cleaning trolley through wireless charging equipment. The design concept of this system is to realize the automatic cleaning of dust on the surface of photovoltaic modules, improve cleaning efficiency and reduce manual intervention. However, in actual applications, since the cleaning trolley and the ferry communicate through 2.4GB wireless signals, and the environment of photovoltaic power stations is complex, the communication signals between robots are easily interfered, resulting in unstable information transmission, which in turn affects the collaborative operation efficiency between the carrier robot and the cleaning robot. In addition, in the face of emergencies such as equipment failure and sudden weather changes, the system often lacks sufficient autonomous decision-making capabilities and is difficult to respond effectively in time, thereby reducing the overall reliability and safety of the system. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention discloses a collaborative operation system and method of a photovoltaic cleaning robot and a carrier robot, aiming to solve the problems raised in the above background technology.

[0005] The present invention adopts the following technical solutions:

[0006] A cooperative operation system of a photovoltaic cleaning robot and a transport robot, comprising:

[0007] The environmental perception module is used to call the cameras, lidars and sensors on the carrier robot and the cleaning robot to collect environmental information of the photovoltaic power station in real time, including terrain, obstacles and weather conditions;

[0008] A collaborative communication module, used to establish a communication channel between the carrier robot and the cleaning robot based on 5G and LoRa communication protocols, and ensure data transmission through an automatic retransmission request mechanism and an error detection and correction code; the collaborative communication module includes a main communication unit and a backup communication unit;

[0009] A communication self-adjusting module is used to monitor the communication quality between the cleaning robot and the carrier robot in real time through a communication status monitoring system during the communication process of the collaborative communication module, including signal strength and bit error rate indicators. According to the monitoring data, the communication self-adjusting module adaptively adjusts the communication parameters, including transmission power and modulation mode, through a machine learning algorithm;

[0010] The path planning module is used to dynamically allocate cleaning tasks to each carrier robot according to the collected data of the environmental perception module, the layout of the photovoltaic power station, the cleaning requirements and the carrying capacity of the carrier robot through a collaborative task allocation algorithm, and plan the optimal path.

[0011] A collaborative obstacle avoidance module is used to dynamically adjust the robot's motion trajectory according to the distance between the obstacle position and the robot using a potential field function-based method based on the real-time collected data of the environment perception module during the path execution process;

[0012] The operation progress tracking module is used to record the operation status of each robot in real time through the operation progress monitoring system during the cleaning operation;

[0013] The output end of the environmental perception module is connected to the input end of the path planning module and the collaborative obstacle avoidance module; the output end of the collaborative communication module is connected to the input end of the communication self-adjustment module, the path planning module, the collaborative obstacle avoidance module and the operation progress tracking module; the output end of the communication self-adjustment module is connected to the input end of the collaborative communication module; the output end of the path planning module is connected to the input end of the collaborative obstacle avoidance module.

[0014] As a further technical solution of the present invention, the main communication unit adopts 5G signals for communication; the backup communication unit adopts LoRa communication protocol for communication; when the main communication unit detects that the signal strength is lower than a preset threshold, it automatically switches to the backup communication unit through a multi-path transmission method.

[0015] As a further technical solution of the present invention, the channel switching method of the main communication unit is: first, the quality of the 5G signal is detected in real time through a signal strength monitoring mechanism; the signal strength monitoring mechanism continuously measures the received 5G signal strength based on the RSSI algorithm, and compares the strength value with a preset threshold; if the detected signal strength is lower than the preset threshold, the main communication unit evaluates the current communication environment through a multi-path transmission method; the multi-path transmission method evaluates the signal quality of different paths through a multi-input multi-output MIMO method, and records the transmission performance of the current path; based on the channel quality evaluation result, the main communication unit determines whether to switch to the backup communication unit through a channel switching algorithm; the channel switching algorithm dynamically adjusts the switching threshold according to the current signal quality and historical data based on an adaptive threshold adjustment mechanism; before channel switching, the main communication unit establishes a connection with the backup communication unit through the LoRa communication protocol, and synchronously transmits the current task status and environmental information to ensure communication continuity; when switching channels, the main communication unit temporarily stores unfinished data packets through a data cache mechanism; after channel switching, the main communication unit reports the channel switching result to the central controller through a heartbeat packet mechanism.

[0016] As a further technical solution of the present invention, the communication status monitoring system includes a signal monitoring module, a data processing module and a bit error rate detection module; the signal monitoring module is used to continuously sample the radio frequency signal in the communication link between the cleaning robot and the carrier robot; the data processing module is used to perform fast Fourier transform and spectrum analysis on the collected digital signal through a real-time signal processing method to extract signal strength, frequency offset and phase distortion parameters; the bit error rate detection module is used to check and count communication data packets; the bit error rate detection module uses a cyclic redundancy check method to verify the integrity of each received data packet, counts the number of error bits, and calculates the bit error rate.

[0017] As a further technical solution of the present invention, the communication self-adjustment module adaptively adjusts the communication parameters by: learning historical communication data through support vector machines and random forests to establish a mapping relationship model between signal strength and bit error rate and communication parameters; in the data learning stage, extracting signal strength, bit error rate, transmission power and modulation mode data from historical communication records through feature engineering to form a training sample set, and ensuring the generalization ability and accuracy of the model through a cross-validation method. After the training is completed, the current signal strength and bit error rate data currently collected are input into the trained mapping relationship model at a regular interval to predict the optimal transmission power and modulation mode, and generate adjustment suggestions, including specific adjustment of the transmission power and modulation mode. numerical value; according to the output of the mapping relationship model, the communication self-adjusting module gradually adjusts the transmission power and modulation mode through the PID controller; after the parameters are adjusted, the communication self-adjusting module also verifies the adjustment effect through the feedback verification mechanism; the feedback verification mechanism adopts a closed-loop control method to monitor the adjusted signal strength and bit error rate in real time, and evaluates the adjustment effect. If the adjustment effect is lower than the preset threshold, the mapping relationship model is triggered to recalculate and adjust the communication parameters. If the adjustment effect is equal to or higher than the preset threshold, the next timing input process is performed; the communication self-adjusting module also records the detailed information of each parameter adjustment through a big data storage method, including the signal strength, bit error rate, transmission power and modulation mode before and after the adjustment.

[0018] As a further technical solution of the present invention, the communication self-adjustment module also defines the command format, parameter range and error handling rules used in the remote configuration process through a remote configuration protocol and instruction set, and sends them to the communication self-adjustment module using a wireless network, thereby supporting operation and maintenance personnel to remotely configure communication parameters.

[0019] As a further technical solution of the present invention, the working method of the collaborative task allocation algorithm is as follows: collect the real-time data of the environmental perception module through data integration, including the layout diagram of the photovoltaic power station, the location of obstacles and the distribution of light intensity; and integrate the collected data into the working environment model through data preprocessing; divide the tasks through a priority allocation mechanism based on the operation requirements and current environmental conditions of the photovoltaic power station; the priority allocation mechanism uses a weight factor to comprehensively consider the cleaning requirements, light intensity distribution, and the battery capacity, load capacity and moving speed of the robot for priority sorting; based on the sorting results, the collaborative task allocation algorithm continuously iterates and optimizes the task allocation plan through a genetic algorithm; in the iterative process, the collaborative task The task allocation algorithm calculates the path length, energy consumption, and time cost indicators according to the current task allocation, and guides the algorithm to converge to a better solution through a heuristic search strategy; after the task allocation plan is determined, the task is assigned to each carrier robot through the wireless network protocol, and the transportation path is planned through the path planning algorithm; the path planning algorithm finds the shortest path from the starting point to the end point through heuristic search, and ensures the global optimality of the path through breadth-first search; during the cleaning execution process, the collaborative task allocation algorithm also continuously receives the execution status information of each carrier robot through the wireless communication network, including position, speed, and remaining power; if the robot execution abnormality is detected, including insufficient power and path obstruction, the task and path are replanned through the reallocation mechanism.

[0020] As a further technical solution of the present invention, the working method of the method based on the potential field function is:

[0021] S1. Acquire the collected environmental data of the environmental perception module in real time, including the position, shape and size of obstacles, and the current position and speed of the robot;

[0022] S2. According to the distance between the obstacle position and the robot, the attraction of the robot to move toward the target position is defined by the gravitational potential field function; the gravitational potential field function adapts to the target tracking requirements in different environments by dynamically adjusting the factor, and the formula expression is:

[0023]

[0024] In formula (1), U att (x) represents the energy value of the attractive potential field; x is the current position of the robot; x goal (t) target position, which changes with time t; k att (t) represents the time-dependent gravity coefficient, which is dynamically adjusted according to environmental changes;

[0025] S3. Define the repulsive force of the robot away from obstacles through a repulsive potential field function; the repulsive potential field function uses the influencing factors of obstacle type and shape to adapt to different types of obstacles, and the formula expression is:

[0026]

[0027] In formula (2), U reple (x) is the energy value of the repulsive potential field; x i (t) represents the position of the i-th obstacle, which changes with time t; r i (t) is the time-dependent influence radius of the ith obstacle, which is dynamically adjusted according to the type and shape of the obstacle; k reple (t) represents the time-dependent repulsion coefficient, which is dynamically adjusted according to environmental changes;

[0028] S4, the attractive potential field U att (x) and the repulsive potential field U reple (x) and we get the total potential field U total (x), guiding the robot’s trajectory. During the synthesis process, the smoothing factor is dynamically adjusted according to the complexity of the environment to balance the influence of attraction and repulsion;

[0029] S5. Calculate the total potential field U by gradient descent method total The gradient of (x) determines the direction of movement of the robot. The formula is:

[0030]

[0031] In formula (3), represents the gradient of the total potential field; σ represents the smoothing factor, which is used to balance the influence of the attractive potential field and the repulsive potential field; represents the gradient of the attractive potential field; represents the gradient of the repulsive potential field;

[0032] S6. According to the calculated gradient, the speed of the robot is adjusted according to the gradient direction through the speed control mechanism to dynamically adjust the motion trajectory of the robot, and the adjusted trajectory is smoothed through the Kalman filter to ensure the continuity and smoothness of the motion trajectory;

[0033] S7. During the path execution process, the robot's position, speed and environmental changes are monitored in real time through the environmental perception module, and the path and speed are dynamically adjusted according to the real-time monitoring results through a closed-loop feedback mechanism.

[0034] As a further technical solution of the present invention, the operation progress monitoring system includes a data acquisition module, a data processing module, a progress analysis module and a visualization display module; the data acquisition module acquires the position, speed, working status and environmental data of each robot in real time through a data interface; and verifies and cleans the collected data through a Kalman filter to remove outliers and noise; the data processing module parses the collected raw data into structured data in JSON format, and after the data is parsed, the parsed data is stored through a distributed database; the data processing module also ensures data synchronization between the central server and each robot through a message queue; the progress analysis module includes a real-time analysis unit, a historical data analysis unit and a prediction analysis unit: the real-time analysis unit performs real-time analysis on the current operation data through a stream processing engine to generate a real-time operation progress report; the historical data analysis unit analyzes the historical operation data through cluster analysis and association rule analysis to generate a historical operation report; the prediction analysis unit predicts future operation progress and potential problems based on historical data and real-time data through time series analysis and regression analysis; the visualization display module displays real-time operation progress and historical operation data through data visualization methods and chart libraries, and provides user data query and filtering through interactive methods.

[0035] As a further technical solution of the present invention, a method for collaborative operation of a photovoltaic cleaning robot and a transport robot comprises the following steps:

[0036] Step 1: Initialize the environment perception module to collect the environmental information around the carrier robot and the cleaning robot, including terrain, obstacles and weather conditions;

[0037] Step 2: Establish a communication channel between the carrier robot and the cleaning robot based on 5G and LoRa communication protocols;

[0038] Step 3: Monitor the communication quality in real time through the communication self-adjustment module, and adaptively adjust the communication parameters, including transmission power and modulation mode, through the machine learning algorithm;

[0039] Step 4: According to the data collected by the environmental perception module and the layout of the photovoltaic power station, cleaning requirements and carrying capacity, the path planning module dynamically allocates cleaning tasks and plans the optimal path;

[0040] Step 5: When performing the cleaning task, dynamically adjust the motion trajectory of the robot through a method based on the potential field function;

[0041] Step 6: During the cleaning task, the operation status of each robot is recorded in real time through the operation progress tracking module, and analyzed and displayed.

[0042] The positive beneficial effects of the present invention are:

[0043] 1. The present invention introduces 5G and LoRa communication protocols to build a more stable and efficient communication channel. Compared with traditional 2.4GB wireless signals, 5G and LoRa have stronger anti-interference capabilities in complex environments, which can effectively reduce the interference of complex terrain and obstacles in photovoltaic power stations on communication signals, and ensure the stability of information transmission between the cleaning robot and the carrier robot. The collaborative communication module includes a main communication unit and a backup communication unit, as well as the application of an automatic retransmission request mechanism and an error detection and correction code, which further enhances the redundancy and fault tolerance of the communication system. Even if a short interruption or error occurs during the communication process, it can be quickly restored and corrected, thereby ensuring the continuity of collaborative operations.

[0044] 2. The communication self-adjustment module monitors the communication quality in real time and adaptively adjusts the communication parameters (such as transmission power and modulation mode) according to indicators such as signal strength and bit error rate. This dynamic adjustment mechanism enables the system to automatically optimize communication performance when facing different environmental conditions, reducing the decline in collaborative work efficiency caused by unstable communication. In addition, the path planning module and the collaborative obstacle avoidance module can dynamically adjust the cleaning tasks and path planning according to real-time environmental information, and use a method based on potential field functions to dynamically adjust the robot's motion trajectory, effectively responding to emergencies such as equipment failures and sudden weather changes, improving the system's flexibility and autonomous decision-making capabilities, and ensuring the safety and efficiency of the operation process.

[0045] 3. The operation progress tracking module records the operation status of each robot in real time through the operation progress monitoring system, providing the system with comprehensive operation progress information. This not only helps managers to understand the operation progress in a timely manner, but also allows them to quickly locate and take measures when problems arise, further improving the overall reliability and operation efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A framework diagram of a cooperative operation system of a photovoltaic cleaning robot and a transport robot according to the present invention;

[0047] Figure 2 A structural framework step diagram of a channel switching method of a main communication unit of the present invention;

[0048] Figure 3 A schematic diagram of the framework of the method for adaptively adjusting communication parameters by the communication self-adjusting module of the present invention;

[0049] Figure 4 This is a working method step diagram of the method based on potential field function of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 are within the scope of protection of the present invention.

[0051] In the present invention, Figure 1-Figure 4 As shown: A cooperative operation system of a photovoltaic cleaning robot and a carrier robot, comprising:

[0052] The environmental perception module is used to call the cameras, lidars and sensors on the carrier robot and the cleaning robot to collect environmental information of the photovoltaic power station in real time, including terrain, obstacles and weather conditions;

[0053] The collaborative communication module is used to establish a communication channel between the carrier robot and the cleaning robot based on 5G and LoRa communication protocols, and ensure data transmission through an automatic retransmission request mechanism and an error detection and correction code; the collaborative communication module includes a main communication unit and a backup communication unit; further, the main communication unit uses 5G signals for communication; the backup communication unit uses the LoRa communication protocol for communication; when the main communication unit detects that the signal strength is lower than a preset threshold, it automatically switches to the backup communication unit through a multipath transmission method; the channel switching method of the main communication unit is: first, the quality of the 5G signal is detected in real time through a signal strength monitoring mechanism; the signal strength monitoring mechanism continuously measures the received 5G signal strength based on the RSSI algorithm, and compares the strength value with a preset threshold; if the detected signal strength is lower than the preset threshold, The main communication unit evaluates the current communication environment through a multi-path transmission method; the multi-path transmission method evaluates the signal quality of different paths through a multi-input multi-output MIMO method, and records the transmission performance of the current path; based on the channel quality evaluation result, the main communication unit determines whether to switch to the backup communication unit through a channel switching algorithm; the channel switching algorithm dynamically adjusts the switching threshold according to the current signal quality and historical data based on an adaptive threshold adjustment mechanism; before channel switching, the main communication unit establishes a connection with the backup communication unit through the LoRa communication protocol, and synchronously transmits the current task status and environmental information to ensure communication continuity; when switching channels, the main communication unit temporarily stores unfinished data packets through a data cache mechanism; after channel switching, the main communication unit reports the channel switching result to the central controller through a heartbeat packet mechanism;

[0054] In the specific implementation, the module uses the high-speed and low-latency characteristics of 5G communication and the long-distance and low-power advantages of LoRa communication to build a highly reliable communication system. The 5G communication unit is responsible for efficient data transmission when the signal is good, while the LoRa communication unit is used as a backup to take over the communication task when the 5G signal is poor to ensure the continuity of data transmission. The strength of the 5G signal is monitored in real time through the RSSI algorithm and compared with the preset threshold. This technical principle can detect the decline in signal quality in time. When the signal strength is lower than the threshold, the system starts multi-path transmission evaluation and selects the best communication path using MIMO technology. If all 5G paths do not meet the requirements, they will automatically switch to LoRa communication. This process is based on the adaptive threshold adjustment mechanism to dynamically adjust the switching threshold to adapt to different communication environments. Before the channel is switched, the current task status and environmental information are synchronized through the LoRa communication protocol to ensure the continuity of information during the switching process. At the same time, the data cache mechanism temporarily stores unfinished data packets at the moment of switching, and continues to transmit after the switching is completed, avoiding data loss.

[0055] In practical applications, the collaborative communication module can significantly improve the communication stability and reliability between the photovoltaic cleaning robot and the carrier robot. In areas with good signals, the 5G communication unit efficiently transmits data and supports real-time collaborative operations between robots. When entering areas with weak signals, the system can automatically switch to LoRa communication to maintain communication continuity and ensure that the robot can continue to receive instructions and feedback status information. In addition, the application of data synchronization and caching mechanisms further enhances the system's fault tolerance and ensures data integrity and continuity even during channel switching.

[0056] The logical operation process of the collaborative communication module is as follows: First, the system monitors the strength of the 5G signal in real time through the RSSI algorithm and compares it with the preset threshold. If the signal strength is lower than the threshold, the multipath transmission evaluation is started and the best communication path is selected using MIMO technology. If all 5G paths do not meet the requirements, the system dynamically adjusts the switching threshold according to the adaptive threshold adjustment mechanism and prepares to switch to LoRa communication. Before switching, the system synchronizes the current task status and environmental information through the LoRa communication protocol, and starts the data cache mechanism to temporarily store unfinished data packets. After the switch is completed, the system continues to transmit the cached data packets and reports the channel switching results to the central controller through the heartbeat packet mechanism to ensure real-time update of the system status.

[0057] Compared with the existing technology, the collaborative communication module significantly improves the stability and reliability of communication through 5G and LoRa dual-mode communication, as well as the adaptive switching mechanism, and can maintain the continuity of communication even in poor signal environments. When the signal is good, the high-speed characteristics of 5G communication are utilized to achieve efficient data transmission and support real-time collaborative operations between robots. Secondly, through the adaptive threshold adjustment mechanism and MIMO technology, the system can dynamically adjust the switching threshold according to the current signal quality and historical data, select the best communication path, and improve the intelligence level of the system. At the same time, the application of data synchronization and caching mechanism enhances the fault tolerance of the system, and can ensure the integrity and continuity of data even during channel switching, thereby improving the robustness of the system.

[0058] A communication self-adjusting module is used to monitor the communication quality between the cleaning robot and the carrier robot in real time through a communication status monitoring system during the communication process of the collaborative communication module, including signal strength and bit error rate indicators. According to the monitoring data, the communication self-adjusting module adaptively adjusts the communication parameters, including transmission power and modulation mode, through a machine learning algorithm;

[0059] Furthermore, the communication status monitoring system includes a signal monitoring module, a data processing module and a bit error rate detection module; the signal monitoring module is used to continuously sample the radio frequency signal in the communication link between the cleaning robot and the carrier robot; the data processing module is used to perform fast Fourier transform and spectrum analysis on the collected digital signal through a real-time signal processing method to extract signal strength, frequency offset and phase distortion parameters; the bit error rate detection module is used to check and count the communication data packets; the bit error rate detection module uses a cyclic redundancy check method to verify the integrity of each received data packet, and counts the number of error bits to calculate the bit error rate;

[0060] In the specific implementation, signal monitoring and real-time sampling use the signal monitoring module to continuously sample the radio frequency signal between the cleaning robot and the carrier robot, and monitor the signal changes in the communication link in real time. This technology can capture key parameters such as signal strength fluctuations, frequency offsets, and phase distortions, and provide data support for subsequent communication parameter adjustments. Through the data processing module, the system performs fast Fourier transform (FFT) and spectrum analysis on the collected digital signals. FFT technology can convert time domain signals into frequency domain signals, thereby extracting the frequency components and energy distribution of the signals. Spectral analysis further reveals the spectral characteristics of the signal, providing an important basis for evaluating signal quality and adjusting communication parameters. The bit error rate detection module uses the cyclic redundancy check (CRC) method to verify the integrity of the received data packets, count the number of error bits, and calculate the bit error rate. CRC technology adds redundant information to the data packet, allowing the receiving end to detect whether errors have occurred during data transmission. The calculation of the bit error rate reflects the overall performance of the communication link and provides feedback for adaptive adjustment of communication parameters.

[0061] In practical applications, the communication status monitoring system can significantly improve the communication quality between the photovoltaic cleaning robot and the carrier robot. The system captures signal changes in the communication link through continuous sampling and real-time monitoring, and extracts key parameters using FFT and spectrum analysis techniques. At the same time, the bit error rate detection module verifies the integrity of the data packet and calculates the bit error rate, providing the system with intuitive communication quality indicators. Based on these data and indicators, the system can adaptively adjust communication parameters such as transmission power and modulation mode to optimize communication performance. This working method not only improves the stability and reliability of communication, but also reduces energy consumption and interference during the communication process.

[0062] In actual work, the communication status monitoring system first continuously samples the radio frequency signal between the cleaning robot and the carrier robot through the signal monitoring module, and passes the sampled data to the data processing module. The data processing module performs FFT and spectrum analysis on the collected digital signal to extract key parameters such as signal strength, frequency offset and phase distortion. At the same time, the bit error rate detection module performs CRC check on the received data packets, counts the number of error bits, and calculates the bit error rate. Based on these data and indicators, the communication self-adjustment module adaptively adjusts the communication parameters through machine learning algorithms, such as increasing the transmission power or changing the modulation mode, to optimize the communication performance. The adjusted communication parameters will be applied to the communication link in real time, and the system will continue to monitor the communication quality and make further adjustments as needed.

[0063] Compared with the existing technology, the communication status monitoring system can monitor the signal changes and bit error rate in the communication link in real time, and adaptively adjust the communication parameters according to these data and indicators. This feature enables the system to quickly respond to changes in the communication environment and maintain the stability and reliability of communication. Secondly, through FFT and spectrum analysis technology, the system can efficiently extract the key parameters of the signal, providing an important basis for evaluating signal quality and adjusting communication parameters. This technology not only improves the efficiency of data processing, but also enhances the intelligence level of the system. By adaptively adjusting the communication parameters, the system can reduce energy consumption and interference while ensuring communication quality. This feature makes the system more energy-saving and environmentally friendly in the collaborative operation of the photovoltaic cleaning robot and the carrier robot. In addition, the various modules of the system have good interfaces and scalability, and can be customized and expanded according to actual needs. At the same time, the adaptive adjustment mechanism of the system also enables it to adapt to different communication environments and application scenarios, and has a wide range of applicability.

[0064] Furthermore, the communication self-adjustment module adaptively adjusts the communication parameters by: learning historical communication data through support vector machines and random forests to establish a mapping relationship model between signal strength and bit error rate and communication parameters; in the data learning stage, extracting signal strength, bit error rate, transmission power and modulation mode data from historical communication records through feature engineering to form a training sample set, and ensuring the generalization ability and accuracy of the model through a cross-validation method. After the training is completed, the current signal strength and bit error rate data currently collected are input into the trained mapping relationship model at regular intervals to predict the optimal transmission power and modulation mode, and generate adjustment suggestions, including adjusting the specific values ​​of the transmission power and modulation mode; according to The output of the mapping relationship model, the communication self-adjustment module gradually adjusts the transmission power and modulation mode through the PID controller; after the parameters are adjusted, the communication self-adjustment module also verifies the adjustment effect through the feedback verification mechanism; the feedback verification mechanism uses a closed-loop control method to monitor the adjusted signal strength and bit error rate in real time, evaluate the adjustment effect, and if the adjustment effect is lower than the preset threshold, the mapping relationship model is triggered to recalculate and adjust the communication parameters. If the adjustment effect is equal to or higher than the preset threshold, the next timed input process is performed; the communication self-adjustment module also records the detailed information of each parameter adjustment through a big data storage method, including the signal strength, bit error rate, transmission power and modulation mode before and after the adjustment. The communication self-adjustment module also defines the command format, parameter range and error handling rules used in the remote configuration process through the remote configuration protocol and instruction set, and uses the wireless network to send it to the communication self-adjustment module to support the operation and maintenance personnel to remotely configure the communication parameters.

[0065] In the specific implementation, the communication self-adjustment module uses machine learning algorithms such as support vector machine (SVM) and random forest (RF) to learn the complex mapping relationship between signal strength, bit error rate and communication parameters (such as transmission power and modulation mode) from historical communication data. SVM maximizes the classification interval by finding the optimal hyperplane, which is suitable for processing high-dimensional data and nonlinear relationships; RF improves prediction accuracy and robustness by building multiple decision trees and synthesizing their output results. These machine learning models can capture the potential laws in the communication environment and provide a scientific basis for adaptively adjusting communication parameters. PID controller is applied for control; PID (proportional-integral-differential) controller is a classic control algorithm used to adjust the control quantity according to the difference between the current state and the target value. In this technical principle, the PID controller gradually adjusts the communication parameters according to the difference between the optimal transmission power and modulation mode predicted by the machine learning model and the current actual value, so as to achieve continuous optimization of communication quality. The introduction of PID controller makes the adjustment of communication parameters smoother and more stable, avoiding communication interruption or performance degradation caused by parameter mutation. The closed-loop feedback verification mechanism evaluates the adjustment effect by real-time monitoring of the adjusted signal strength and bit error rate, and decides whether to recalculate and adjust the communication parameters based on the evaluation results. This mechanism ensures the accuracy and effectiveness of the communication parameter adjustment process, allowing the system to quickly respond to changes in the communication environment and maintain stable communication quality. Big data storage technology is used to record detailed information on each parameter adjustment, including signal strength, bit error rate, transmission power, and modulation mode before and after adjustment, providing a rich data source for subsequent data analysis and performance optimization. The remote configuration protocol and instruction set define the command format, parameter range, and error handling rules used in the remote configuration process, allowing operation and maintenance personnel to remotely configure communication parameters through wireless networks, improving the flexibility and maintainability of the system.

[0066] In actual applications, the communication self-adjustment module learns the mapping relationship between signal strength, bit error rate and communication parameters from historical communication data through a machine learning model, and predicts the optimal transmission power and modulation mode based on these relationships. Then, the PID controller gradually adjusts the communication parameters according to the prediction results to optimize the communication quality. At the same time, the closed-loop feedback verification mechanism monitors the adjusted communication quality in real time and decides whether to readjust the parameters based on the evaluation results. In addition, big data storage technology records the detailed information of each parameter adjustment, providing data support for subsequent performance analysis and optimization. The remote configuration function enables operation and maintenance personnel to easily adjust communication parameters remotely, improving the flexibility and maintainability of the system. This working method not only improves the stability and reliability of communication, but also reduces operation and maintenance costs and time costs.

[0067] In the specific implementation, the working process of the communication self-adjustment module is as follows: first, extract data such as signal strength, bit error rate, transmission power and modulation mode from historical communication records to form a training sample set, and use machine learning algorithms such as SVM and RF to train the mapping relationship model. Then, collect the current signal strength and bit error rate data regularly, input them into the trained mapping relationship model, and predict the optimal transmission power and modulation mode. Then, the PID controller gradually adjusts the communication parameters according to the prediction results. During the adjustment process, the closed-loop feedback verification mechanism monitors the adjusted communication quality in real time and decides whether to readjust the parameters based on the evaluation results. If the adjustment effect is lower than the preset threshold, the mapping relationship model is triggered to recalculate and adjust the communication parameters; if the adjustment effect is equal to or higher than the preset threshold, the next timed input process is performed. At the same time, big data storage technology records the detailed information of each parameter adjustment. In addition, operation and maintenance personnel can remotely configure communication parameters through remote configuration protocols and instruction sets, which improves the flexibility and maintainability of the system.

[0068] Compared with the existing technology, the communication self-adjustment module realizes the intelligent and automatic adjustment of communication parameters through the application of machine learning models and PID controllers, improving the system's adaptability and response speed. At the same time, the mapping relationship model is constructed using machine learning algorithms such as SVM and RF to improve the accuracy of prediction; the introduction of PID controller makes parameter adjustment smoother and more stable, avoiding the risk of communication interruption or performance degradation. Secondly, the closed-loop feedback verification mechanism ensures the accuracy and effectiveness of the communication parameter adjustment process; big data storage technology provides data support for subsequent performance analysis and optimization, enabling the system to continuously optimize communication quality. In addition, the definition of remote configuration protocols and instruction sets enables operation and maintenance personnel to easily adjust communication parameters remotely, improving the flexibility and maintainability of the system. This feature is particularly important in the collaborative operation of photovoltaic cleaning robots and carrier robots, because operation and maintenance personnel may need to monitor and maintain the system in different geographical locations.

[0069] The path planning module is used to dynamically allocate cleaning tasks to each carrier robot according to the collected data of the environmental perception module, the layout of the photovoltaic power station, the cleaning requirements and the carrying capacity of the carrier robot through a collaborative task allocation algorithm, and plan the optimal path.

[0070] Furthermore, the working method of the collaborative task allocation algorithm is as follows: collect real-time data of the environmental perception module through data integration, including the layout diagram of the photovoltaic power station, obstacle locations and light intensity distribution; and integrate the collected data into the working environment model through data preprocessing; divide tasks through a priority allocation mechanism based on the operating needs of the photovoltaic power station and current environmental conditions; the priority allocation mechanism uses weight factors to comprehensively consider cleaning needs, light intensity distribution, and robot battery capacity, load capacity and moving speed for priority sorting; based on the sorting results, the collaborative task allocation algorithm continuously iterates and optimizes the task allocation plan through a genetic algorithm; during the iteration process, the collaborative task allocation algorithm According to the current task allocation, the path length, energy consumption and time cost indicators are calculated, and the algorithm is guided to converge to a better solution through a heuristic search strategy; after the task allocation plan is determined, the task is assigned to each carrier robot through the wireless network protocol, and the transportation path is planned through the path planning algorithm; the path planning algorithm finds the shortest path from the starting point to the end point through heuristic search, and ensures the global optimality of the path through breadth-first search; during the cleaning execution process, the collaborative task allocation algorithm also continuously receives the execution status information of each carrier robot through the wireless communication network, including position, speed, and remaining power; if the robot execution abnormality is detected, including insufficient power and path obstruction, the task and path are replanned through the reallocation mechanism.

[0071] In the specific implementation, the real-time data integration and preprocessing technology collects real-time data from the environmental perception module, including the layout of the photovoltaic power station, the location of obstacles, and the distribution of light intensity, and integrates these data into the operating environment model after preprocessing, providing an accurate description of the environment for subsequent task allocation and path planning. Secondly, the priority allocation mechanism comprehensively considers multiple factors such as cleaning requirements, light intensity distribution, and the battery capacity, load capacity, and moving speed of the robot, and quantifies these factors through weight factors to determine the priority order of tasks. Thirdly, the genetic algorithm optimization technology simulates natural selection and genetic mechanisms, continuously iteratively searches the solution space, and finds the optimal task allocation solution, so that the system can find an approximate optimal solution in complex task allocation problems. In addition, the heuristic search strategy uses certain characteristics or empirical knowledge of the problem to accelerate the search process and improve the search efficiency, and guide the algorithm to converge to a better solution. Finally, the path planning algorithm uses a heuristic search method to find the shortest path from the starting point to the end point, and ensures the global optimality of the path through breadth-first search, and plans the optimal path for each carrier robot.

[0072] In practical applications, the collaborative task allocation algorithm collects data from the environmental perception module in real time, and dynamically adjusts the task allocation and path planning scheme based on this data. This real-time data-driven approach enables the system to cope with the complex and changeable photovoltaic power station environment and improves work efficiency. Secondly, the algorithm ensures that the system can prioritize important and urgent tasks and improve work quality through priority sorting and task allocation. At the same time, the iterative optimization of the genetic algorithm enables the system to find an approximate optimal solution in complex task allocation problems, further improving work efficiency. In addition, the guarantee of path planning and global optimality ensures that the carrier robot can move along the optimal path when performing cleaning tasks, reducing energy consumption and time costs. Finally, the algorithm continuously receives the execution status information of each carrier robot through the wireless communication network, and detects whether the robot is abnormal in execution based on this information. If an abnormality is detected, the task and path are replanned through the reallocation mechanism to ensure the smooth progress of the operation. These working methods not only improve the system's work efficiency and work quality, but also enhance the system's flexibility and robustness, enabling the system to adapt to the operation needs of photovoltaic power stations of different scales and types.

[0073] Compared with the existing technology, the collaborative task allocation algorithm has significant advantages and characteristics. It can realize efficient and accurate task allocation and path planning, improve work efficiency and work quality. At the same time, the algorithm is flexible and robust, and can adapt to the operation needs of photovoltaic power stations of different scales and types. In addition, the guarantee of global optimality and the abnormal detection and rapid response mechanism further enhance the stability and reliability of the system. These advantages and characteristics make the collaborative task allocation algorithm have broad application prospects and important practical value in the collaborative operation control system of photovoltaic cleaning robots and carrier robots.

[0074] In the specific implementation, the hardware working environment of the collaborative task allocation algorithm mainly includes the following parts:

[0075] Photovoltaic power station environment: including photovoltaic panel arrays, ground obstacles (such as trees, buildings), light intensity distribution, etc.

[0076] Environmental perception module: includes cameras, lidar, infrared sensors, etc., used to collect environmental data of photovoltaic power stations in real time.

[0077] Central control unit: a high-performance computer or embedded system used to run collaborative task allocation algorithms and path planning algorithms.

[0078] Communication network: including wireless network and wired network, used to realize data communication between the central control unit and each transport robot.

[0079] Carrying robot: equipped with cleaning equipment and sensors, used to perform cleaning tasks and feedback execution status information.

[0080] Based on the above hardware components, an experiment was designed to verify the effect of the collaborative task allocation algorithm. The experiment selected a set of identical hardware environments and applied the collaborative task allocation algorithm (Group A) and the greedy algorithm (Group B) for comparative experiments. The greedy algorithm is an algorithm that selects the current optimal solution in each step in the hope of eventually obtaining the global optimal solution. Specifically, when allocating tasks, the greedy algorithm selects the current optimal task for allocation based on the current environmental information and robot status, without considering possible changes in the future.

[0081] Both experiments were conducted in the same photovoltaic power station environment, with the same obstacle distribution and the same light intensity distribution. During the experiment, the cleaning efficiency (the area of ​​photovoltaic panels cleaned per unit time), energy consumption (the electric energy consumed to complete the cleaning task) and path length (the moving distance of the transport robot from the starting point to the end point) of each experiment were recorded. Each experiment was conducted five times, and the average value was taken as the final result; the experimental results are shown in Table 1:

[0082] Table 1 Experimental data record table

[0083]

[0084]

[0085] By comparing the data in the experimental record table, it can be seen that the cleaning efficiency of the collaborative task allocation algorithm (Group A) is significantly higher than that of the greedy algorithm (Group B), while the energy consumption is lower and the path length is shorter. This shows that the collaborative task allocation algorithm can more effectively allocate cleaning tasks and optimize the moving path of the carrier robot, thereby improving cleaning efficiency and reducing energy consumption.

[0086] A collaborative obstacle avoidance module is used to dynamically adjust the robot's motion trajectory according to the distance between the obstacle position and the robot using a potential field function-based method based on the real-time collected data of the environment perception module during the path execution process;

[0087] Furthermore, the working method of the method based on the potential field function is:

[0088] S1. Acquire the collected environmental data of the environmental perception module in real time, including the position, shape and size of obstacles, and the current position and speed of the robot;

[0089] S2. According to the distance between the obstacle position and the robot, the attraction of the robot to move toward the target position is defined by the gravitational potential field function; the gravitational potential field function adapts to the target tracking requirements in different environments by dynamically adjusting the factor, and the formula expression is:

[0090]

[0091] In formula (1), U att (x) represents the energy value of the attractive potential field; x is the current position of the robot; x goal (t) target position, which changes with time t; k att (t) represents the time-dependent gravity coefficient, which is dynamically adjusted according to environmental changes;

[0092] S3. Define the repulsive force of the robot away from obstacles through a repulsive potential field function; the repulsive potential field function uses the influencing factors of obstacle type and shape to adapt to different types of obstacles, and the formula expression is:

[0093]

[0094] In formula (2), U reple (x) is the energy value of the repulsive potential field; x i (t) represents the position of the i-th obstacle, which changes with time t; r i (t) is the time-dependent influence radius of the ith obstacle, which is dynamically adjusted according to the type and shape of the obstacle; k reple (t) represents the time-dependent repulsion coefficient, which is dynamically adjusted according to environmental changes;

[0095] S4, the attractive potential field U att (x) and the repulsive potential field U reple (x) and we get the total potential field U total (x), guiding the robot’s trajectory. During the synthesis process, the smoothing factor is dynamically adjusted according to the complexity of the environment to balance the influence of attraction and repulsion;

[0096] S5. Calculate the total potential field U by gradient descent method total The gradient of (x) determines the direction of movement of the robot. The formula is:

[0097]

[0098] In formula (3), represents the gradient of the total potential field; σ represents the smoothing factor, which is used to balance the influence of the attractive potential field and the repulsive potential field; represents the gradient of the attractive potential field; represents the gradient of the repulsive potential field;

[0099] S6. According to the calculated gradient, the speed of the robot is adjusted according to the gradient direction through the speed control mechanism to dynamically adjust the motion trajectory of the robot, and the adjusted trajectory is smoothed through the Kalman filter to ensure the continuity and smoothness of the motion trajectory;

[0100] S7. During the path execution process, the robot's position, speed and environmental changes are monitored in real time through the environmental perception module, and the path and speed are dynamically adjusted according to the real-time monitoring results through a closed-loop feedback mechanism.

[0101] In the specific implementation, the method based on potential field function guides the movement of the robot by constructing a virtual potential field. Specifically, this method combines the concepts of gravitational potential field and repulsive potential field. The gravitational potential field simulates the attraction of the robot moving toward the target position, and its size is related to the distance from the current position of the robot to the target position, and it adapts to the target tracking requirements in different environments by dynamically adjusting factors. The repulsive potential field simulates the repulsive force of the robot away from obstacles, and its size is related to the distance from the robot to the obstacle and the type and shape of the obstacle. It also adapts to different types of obstacles by dynamically adjusting factors. The superposition of these two potential fields forms a total potential field, and the gradient of the total potential field determines the direction of movement of the robot. In addition, in order to balance the influence of attraction and repulsion, a smoothing factor is introduced, which is dynamically adjusted according to the complexity of the environment. Through the comprehensive application of these principles, the method based on potential field function can guide the robot to achieve autonomous obstacle avoidance and path planning in complex environments.

[0102] In practical applications, the method based on potential field function can obtain the collected data of the environmental perception module in real time, including the position, shape and size of obstacles, as well as the current position and speed of the robot, which provides a basis for subsequent potential field calculation. Secondly, by constructing gravitational potential field and repulsive potential field, and dynamically adjusting related factors to adapt to different environments and obstacles, this method can generate a total potential field that conforms to the actual situation, thereby guiding the robot's motion trajectory. Furthermore, the gradient of the total potential field is calculated by the gradient descent method to determine the robot's motion direction, and the robot's speed is adjusted by the speed control mechanism, realizing precise control of the robot's motion. In addition, the method also smoothes the adjusted trajectory through the Kalman filter, ensuring the continuity and smoothness of the motion trajectory, and improving the robot's motion stability and safety. These working methods enable the method based on potential field function to significantly improve the robot's obstacle avoidance ability and path planning efficiency in the collaborative operation control system of the photovoltaic cleaning robot and the carrier robot, reduce the need for manual intervention, and improve the operation efficiency and quality.

[0103] Compared with the existing technology, the method based on potential field function can obtain environmental data in real time and construct a potential field model that conforms to the actual situation, thereby realizing precise control of robot motion. Secondly, through the application of dynamic adjustment factors and smoothing factors, this method can adapt to the needs of different environments and obstacle types, and improve the flexibility and adaptability of the system. Furthermore, this method realizes the precise calculation of the robot's motion direction and dynamic adjustment of the speed through the gradient descent method and speed control mechanism, which improves the robot's obstacle avoidance ability and path planning efficiency. In addition, this method also smoothes the trajectory through the Kalman filter, ensures the continuity and smoothness of the motion trajectory, and improves the robot's motion stability and safety. Finally, this method can be combined with the environmental perception module and the closed-loop feedback mechanism to form a complete control system, realizing real-time and precise control of the robot's motion. These advantages and characteristics make the method based on potential field function have broad application prospects and important practical value in the collaborative operation control system of photovoltaic cleaning robots and carrier robots.

[0104] In a specific implementation, the hardware working environment of the method based on the potential field function includes:

[0105] Photovoltaic cleaning robot: equipped with cameras, lidar and sensors to collect environmental information in real time.

[0106] Transport robot: It has a high load capacity and moving speed and is used to transport the cleaning robot to the designated location.

[0107] Central controller: a high-performance computing device responsible for data processing, task allocation, and path planning.

[0108] Wireless communication network: 5G and LoRa communication protocols ensure real-time transmission of data.

[0109] Environmental perception module: includes cameras, lidar and sensors, which are used to collect the layout of the photovoltaic power station, obstacle locations and light intensity distribution in real time.

[0110] Data storage and processing equipment: used to store and process real-time collected data, supporting big data analysis and machine learning algorithms.

[0111] Based on the above hardware environment, a comparative experiment was designed to verify the application effect of the method based on the potential field function. The experiment was divided into Group A and Group B. Group A applied the method based on the potential field function, and Group B applied the traditional static path planning algorithm. In this experiment, the static path planning algorithm pre-planned the path based on the known environmental information before the task started, without considering the dynamic changes during the path execution. Specifically, the static path planning algorithm pre-calculates the shortest path from the starting point to the end point through the Dijkstra algorithm or the A* algorithm, and then performs the task according to the predetermined path. Both groups of experiments were conducted under the same experimental conditions, and each group conducted five experiments. The main indicators of the experiment include task completion time, path length, obstacle avoidance success rate and task success rate. The experimental data comparison table is shown in Table 2:

[0112] Table 2 Experimental data record table

[0113]

[0114] As shown in Table 2, the average task completion time of Group A is 36.8 minutes, while the average task completion time of Group B is 46.6 minutes. The task completion time of Group A is significantly shorter, which improves work efficiency. The average path length of Group A is 818 meters, while the average path length of Group B is 1018 meters. The path length of Group A is significantly shorter, which optimizes path planning. The average obstacle avoidance success rate of Group A is 97.6%, while the average obstacle avoidance success rate of Group B is 84.6%. The obstacle avoidance success rate of Group A is higher, which improves the robot's obstacle avoidance ability. The average task success rate of Group A is 94.6%, while the average task success rate of Group B is 84.6%. The task success rate of Group A is higher, which improves the reliability and stability of the task.

[0115] The operation progress tracking module is used to record the operation status of each robot in real time through the operation progress monitoring system during the cleaning operation; further, the operation progress monitoring system includes a data acquisition module, a data processing module, a progress analysis module and a visual display module; the data acquisition module obtains the position, speed, working status and environmental data of each robot in real time through the data interface; and verifies and cleans the collected data through the Kalman filter to remove abnormal values ​​and noise; the data processing module parses the collected raw data into structured data in JSON format, and after the data is parsed, the parsed data is stored in a distributed database; the data processing module also ensures the central service through a message queue The data between the server and each robot is synchronized; the progress analysis module includes a real-time analysis unit, a historical data analysis unit and a predictive analysis unit: the real-time analysis unit performs real-time analysis on the current job data through a stream processing engine to generate a real-time job progress report; the historical data analysis unit analyzes the historical job data through clustering analysis and association rule analysis to generate a historical job report; the predictive analysis unit predicts future job progress and potential problems based on historical data and real-time data through time series analysis and regression analysis; the visualization display module displays real-time job progress and historical job data through data visualization methods and chart libraries, and provides user data query and filtering through interactive methods.

[0116] In the specific implementation, the data acquisition module uses data interface technology to achieve real-time communication with each robot, and can efficiently and accurately obtain position, speed, working status and environmental data. These data are the basis for subsequent analysis and decision-making. Secondly, the Kalman filter principle is applied to the data verification and cleaning process. Through the prediction and update mechanism, outliers and noise are effectively removed, and the accuracy and reliability of the data are improved. Furthermore, the data processing module uses JSON format to parse the original data, converts it into structured data, and uses a distributed database for storage, ensuring efficient management and access of the data. In addition, the message queue mechanism ensures data synchronization between the central server and each robot, and realizes real-time update and sharing of data. Finally, the progress analysis module comprehensively applies technical principles such as stream processing engine, cluster analysis, association rule analysis, time series analysis and regression analysis, conducts in-depth analysis of the operation data, generates real-time operation progress reports, historical operation reports and future operation progress forecasts, and provides strong support for decision-making.

[0117] In the specific implementation, the operation progress monitoring system can record the operation status of each robot in real time, including position, speed, working status and environmental data, etc., providing managers with comprehensive operation information. Secondly, the Kalman filter is used to verify and clean the data, effectively avoiding the influence of outliers and noise on the analysis results, and improving the accuracy and reliability of the data. Furthermore, the system uses a distributed database to store data, ensuring efficient management and access of data, and providing strong support for subsequent in-depth analysis. In addition, the progress analysis module can generate operation progress reports in real time, helping managers to understand the operation progress and potential problems in a timely manner, and providing timely and accurate basis for decision-making. Finally, through the visualization display module, the system can present complex operation data in an intuitive and easy-to-understand way, improving the decision-making efficiency and accuracy of managers. These working methods not only improve the overall performance and reliability of the system, but also significantly improve the efficiency and quality of the collaborative operation of photovoltaic cleaning robots and carrier robots.

[0118] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that these specific embodiments are only illustrative, and those skilled in the art may omit, replace, and change the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, merging the above method steps so as to perform substantially the same functions in substantially the same manner to achieve substantially the same results is within the scope of the present invention. Therefore, the scope of the present invention is limited only by the appended claims.

Claims

1. A cooperative operation system of a photovoltaic cleaning robot and a transport robot, characterized in that: include: The environmental perception module is used to call the cameras, lidars and sensors on the carrier robot and the cleaning robot to collect environmental information of the photovoltaic power station in real time, including terrain, obstacles and weather conditions; A collaborative communication module, used to establish a communication channel between the carrier robot and the cleaning robot based on 5G and LoRa communication protocols, and ensure data transmission through an automatic retransmission request mechanism and an error detection and correction code; the collaborative communication module includes a main communication unit and a backup communication unit; A communication self-adjusting module is used to monitor the communication quality between the cleaning robot and the carrier robot in real time through a communication status monitoring system during the communication process of the collaborative communication module, including signal strength and bit error rate indicators. According to the monitoring data, the communication self-adjusting module adaptively adjusts the communication parameters, including transmission power and modulation mode, through a machine learning algorithm; The path planning module is used to dynamically allocate cleaning tasks to each carrier robot according to the collected data of the environmental perception module, the layout of the photovoltaic power station, the cleaning requirements and the carrying capacity of the carrier robot through a collaborative task allocation algorithm, and plan the optimal path. A collaborative obstacle avoidance module is used to dynamically adjust the robot's motion trajectory according to the distance between the obstacle position and the robot using a potential field function-based method based on the real-time collected data of the environment perception module during the path execution process; The operation progress tracking module is used to record the operation status of each robot in real time through the operation progress monitoring system during the cleaning operation; The output end of the environmental perception module is connected to the input end of the path planning module and the collaborative obstacle avoidance module; the output end of the collaborative communication module is connected to the input end of the communication self-adjustment module, the path planning module, the collaborative obstacle avoidance module and the operation progress tracking module; the output end of the communication self-adjustment module is connected to the input end of the collaborative communication module; the output end of the path planning module is connected to the input end of the collaborative obstacle avoidance module.

2. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The main communication unit uses 5G signals for communication; the backup communication unit uses the LoRa communication protocol for communication; when the main communication unit detects that the signal strength is lower than a preset threshold, it automatically switches to the backup communication unit through a multi-path transmission method.

3. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 2, characterized in that: The channel switching method of the main communication unit is as follows: first, the quality of the 5G signal is detected in real time through a signal strength monitoring mechanism; the signal strength monitoring mechanism continuously measures the received 5G signal strength based on the RSSI algorithm, and compares the strength value with a preset threshold; if the detected signal strength is lower than the preset threshold, the main communication unit evaluates the current communication environment through a multi-path transmission method; the multi-path transmission method evaluates the signal quality of different paths through a multi-input multi-output MIMO method, and records the transmission performance of the current path; based on the channel quality evaluation result, the main communication unit determines whether to switch to the backup communication unit through a channel switching algorithm; the channel switching algorithm dynamically adjusts the switching threshold according to the current signal quality and historical data based on an adaptive threshold adjustment mechanism; before channel switching, the main communication unit establishes a connection with the backup communication unit through the LoRa communication protocol, and synchronously transmits the current task status and environmental information to ensure communication continuity; when switching channels, the main communication unit temporarily stores unfinished data packets through a data cache mechanism; after channel switching, the main communication unit reports the channel switching result to the central controller through a heartbeat packet mechanism.

4. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The communication status monitoring system includes a signal monitoring module, a data processing module and a bit error rate detection module; the signal monitoring module is used to continuously sample the radio frequency signal in the communication link between the cleaning robot and the carrier robot; the data processing module is used to perform fast Fourier transform and spectrum analysis on the collected digital signal through a real-time signal processing method to extract signal strength, frequency offset and phase distortion parameters; the bit error rate detection module is used to verify and count the communication data packets; The bit error rate detection module uses a cyclic redundancy check method to verify the integrity of each received data packet, counts the number of error bits, and calculates the bit error rate.

5. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The communication self-adjustment module adaptively adjusts the communication parameters in the following method: historical communication data is learned through support vector machines and random forests to establish a mapping relationship model between signal strength and bit error rate and communication parameters; in the data learning stage, signal strength, bit error rate, transmission power and modulation mode data are extracted from historical communication records through feature engineering to form a training sample set, and the generalization ability and accuracy of the model are ensured through a cross-validation method. After the training is completed, the current signal strength and bit error rate data currently collected are input into the trained mapping relationship model at a regular interval to predict the optimal transmission power and modulation mode, and generate adjustment suggestions, including specific values ​​for adjusting the transmission power and modulation mode; according to the The output of the mapping relationship model, the communication self-adjustment module gradually adjusts the transmission power and modulation mode through the PID controller; after the parameters are adjusted, the communication self-adjustment module also verifies the adjustment effect through the feedback verification mechanism; the feedback verification mechanism adopts a closed-loop control method to monitor the adjusted signal strength and bit error rate in real time, evaluate the adjustment effect, if the adjustment effect is lower than the preset threshold, the mapping relationship model is triggered to recalculate and adjust the communication parameters, if the adjustment effect is equal to or higher than the preset threshold, the next timing input process is performed; the communication self-adjustment module also records the detailed information of each parameter adjustment through a big data storage method, including the signal strength, bit error rate, transmission power and modulation mode before and after the adjustment.

6. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 5, characterized in that: The communication self-adjusting module also defines the command format, parameter range and error handling rules used in the remote configuration process through the remote configuration protocol and instruction set, and sends them to the communication self-adjusting module using a wireless network, supporting operation and maintenance personnel to remotely configure communication parameters.

7. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The working method of the collaborative task allocation algorithm is as follows: collecting real-time data of the environment perception module through data integration, including the layout diagram of the photovoltaic power station, obstacle locations and light intensity distribution; and integrating the collected data into the working environment model through data preprocessing; dividing tasks through a priority allocation mechanism based on the operation requirements of the photovoltaic power station and the current environmental conditions; the priority allocation mechanism uses a weight factor to comprehensively consider the cleaning requirements, light intensity distribution, and the battery capacity, load capacity and moving speed of the robot for priority sorting; Based on the sorting results, the collaborative task allocation algorithm continuously iterates and optimizes the task allocation plan through a genetic algorithm; during the iteration process, the collaborative task allocation algorithm calculates the path length, energy consumption, and time cost indicators according to the current task allocation, and guides the algorithm to converge to a better solution through a heuristic search strategy; after the task allocation plan is determined, the task is assigned to each carrier robot through a wireless network protocol, and the carrier path is planned through a path planning algorithm; the path planning algorithm finds the shortest path from the starting point to the end point through a heuristic search, and ensures the global optimality of the path through a breadth-first search; during the cleaning execution process, the collaborative task allocation algorithm also continuously receives the execution status information of each carrier robot through the wireless communication network, including position, speed, and remaining power; if an abnormal robot execution is detected, including insufficient power and blocked path, the task and path are replanned through a reallocation mechanism.

8. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The working method of the method based on potential field function is: S1. Acquire the collected environmental data of the environmental perception module in real time, including the position, shape and size of obstacles, and the current position and speed of the robot; S2. According to the distance between the obstacle position and the robot, the attraction of the robot to move toward the target position is defined by the gravitational potential field function; the gravitational potential field function adapts to the target tracking requirements in different environments by dynamically adjusting the factor, and the formula expression is: In formula (1), U att (x) represents the energy value of the attractive potential field; x is the current position of the robot; x goal (t) target position, which changes with time t; k att (t) represents the time-dependent gravity coefficient, which is dynamically adjusted according to environmental changes; S3. Define the repulsive force of the robot away from obstacles through a repulsive potential field function; the repulsive potential field function uses the influencing factors of obstacle type and shape to adapt to different types of obstacles, and the formula expression is: In formula (2), U reple (x) is the energy value of the repulsive potential field; x i (t) represents the position of the i-th obstacle, which changes with time t; r i (t) is the time-dependent influence radius of the ith obstacle, which is dynamically adjusted according to the type and shape of the obstacle; k reple (t) represents the time-dependent repulsion coefficient, which is dynamically adjusted according to environmental changes; S4, the attractive potential field U att (x) and the repulsive potential field U reple (x) and we get the total potential field U total (x), guiding the robot’s trajectory. During the synthesis process, the smoothing factor is dynamically adjusted according to the complexity of the environment to balance the influence of attraction and repulsion; S5. Calculate the total potential field U by gradient descent method total The gradient of (x) determines the direction of movement of the robot. The formula is: In formula (3), represents the gradient of the total potential field; σ represents the smoothing factor, which is used to balance the influence of the attractive potential field and the repulsive potential field; represents the gradient of the attractive potential field; represents the gradient of the repulsive potential field; S6. According to the calculated gradient, the speed of the robot is adjusted according to the gradient direction through the speed control mechanism to dynamically adjust the motion trajectory of the robot, and the adjusted trajectory is smoothed through the Kalman filter to ensure the continuity and smoothness of the motion trajectory; S7. During the path execution process, the robot's position, speed and environmental changes are monitored in real time through the environmental perception module, and the path and speed are dynamically adjusted according to the real-time monitoring results through a closed-loop feedback mechanism.

9. The cooperative operation system of a photovoltaic cleaning robot and a transport robot according to claim 1, characterized in that: The operation progress monitoring system includes a data acquisition module, a data processing module, a progress analysis module and a visualization display module; the data acquisition module acquires the position, speed, working status and environmental data of each robot in real time through a data interface; and verifies and cleans the collected data through a Kalman filter to remove outliers and noise; the data processing module parses the collected raw data into structured data in JSON format, and after data parsing, stores the parsed data through a distributed database; the data processing module also ensures data synchronization between the central server and each robot through a message queue; the progress analysis module includes a real-time analysis unit, a historical data analysis unit and a prediction analysis unit: the real-time analysis unit performs real-time analysis on the current operation data through a stream processing engine to generate a real-time operation progress report; the historical data analysis unit analyzes the historical operation data through clustering analysis and association rule analysis to generate a historical operation report; the prediction analysis unit predicts future operation progress and potential problems based on historical data and real-time data through time series analysis and regression analysis; The visualization display module displays real-time job progress and historical job data through data visualization methods and chart libraries, and provides user data query and filtering through interactive methods.

10. A collaborative operation method of a photovoltaic cleaning robot and a transport robot, characterized in that: A cooperative operation system of a photovoltaic cleaning robot and a transport robot as described in any one of claims 1 to 9 comprises the following steps: Step 1: Initialize the environment perception module to collect the environmental information around the carrier robot and the cleaning robot, including terrain, obstacles and weather conditions; Step 2: Establish a communication channel between the carrier robot and the cleaning robot based on 5G and LoRa communication protocols; Step 3: Monitor the communication quality in real time through the communication self-adjustment module, and adaptively adjust the communication parameters, including transmission power and modulation mode, through the machine learning algorithm; Step 4: According to the data collected by the environmental perception module and the layout of the photovoltaic power station, cleaning requirements and carrying capacity, the path planning module dynamically allocates cleaning tasks and plans the optimal path; Step 5: When performing the cleaning task, dynamically adjust the motion trajectory of the robot through a method based on the potential field function; Step 6: During the cleaning task, the operation status of each robot is recorded in real time through the operation progress tracking module, and analyzed and displayed.

Images