Multi-robot cooperative charging system

Abstract

This invention discloses a multi-robot collaborative charging system. The charging system includes a charging main body, a positioning unit, a computing unit, a scheduling unit, and a linkage control unit. The charging main body provides docking spaces and power replenishment for mobile robots. The positioning unit receives satellite positioning signals and calculates and outputs the position signals of the mobile robots. The computing unit receives the position signals output by the positioning unit, predicts the trajectory of the mobile robots, and outputs coordination signals to the corresponding mobile robots to coordinate multi-robot collaborative operations. The scheduling unit receives the coordination signals from the computing unit and generates scheduling signals for multi-robot collaborative operations. The linkage control unit receives the coordination signals from the computing unit and outputs linkage signals to the corresponding IoT devices. The technical solution of this application can effectively improve the efficiency of multi-robot collaborative operations.

Description

Technical Field

[0001] This invention relates to the field of charging technology for operational robots, and more specifically to a multi-robot collaborative charging system. Background Technology

[0002] With the development of outdoor mobile robot technology, mobile robots for tasks such as lawn mowing and yard inspection have gradually achieved large-scale collaborative operations. However, current robot charging systems are still limited to basic power replenishment and cannot meet the needs of multiple robots operating simultaneously. Specifically, the lack of a unified scheduling and coordination mechanism during multi-robot operations means that each robot independently executes preset work instructions, which can easily lead to problems such as conflicting work paths and delays in work processes, resulting in a significant reduction in overall work efficiency. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a multi-robot collaborative charging system that can effectively improve the efficiency of multi-robot collaborative operations.

[0004] This application provides a multi-robot collaborative charging system, the charging system comprising:

[0005] A charging unit, which provides a docking bay and power supply for the mobile robot.

[0006] The positioning unit is used to receive satellite positioning signals and calculate and output the position signal of the mobile operation robot;

[0007] The computing unit is used to receive the position signal output by the positioning unit, predict the trajectory of the mobile robot, and output a coordination signal to the corresponding mobile robot to coordinate multi-robot collaborative operation.

[0008] A scheduling unit is used to receive coordination signals from the computing unit and generate scheduling signals for multi-machine collaborative operations.

[0009] The linkage control unit is used to receive coordination signals from the computing unit and output linkage signals to the corresponding Internet of Things devices.

[0010] In one aspect, the positioning unit integrates satellite observation data and inertial sensing data to perform differential calculations, outputs the position signal of the mobile robot, and weights different satellite observations with signal-to-noise ratio as the weight, corrects the observation residuals according to the weights, and then outputs the position signal.

[0011] In one aspect, the coordination signal output by the computing unit includes job cost optimization parameters and trajectory prediction parameters;

[0012] The operation cost optimization parameters are defined with the objective function J = Σdi + λ1 * ΣPconflict(i, j) + λ2 * ΣDelayi, and the trajectory prediction parameters are defined with Tpred = γ * T. L +(1-γ)*T K The weighted fusion strategy is defined as follows: di is the remaining path length of mobile robot i, Pconflict(i,j) is the path conflict probability between mobile robots i and j, Delayi is the operation time delay of mobile robot i, λ1 and λ2 are preset weight coefficients, λ1 > λ2, and T... L T represents the long-term trajectory prediction result of the neural network model. K The result is the short-term trajectory prediction of the filtering algorithm, where γ is the fusion coefficient and i and j are the number of robots.

[0013] In one aspect, when the computing unit predicts the trajectory of the mobile robot,

[0014] The short-term trajectory prediction value T is obtained through the Kalman filter algorithm. K =Φ*T K-1 +E*uk+wk;

[0015] The long-term trajectory prediction value T is obtained through the long short-term memory network model. L =LSTM(P t-n P t-n+1 ..., Pt);

[0016] The fused trajectory prediction results are output to the coordination signal according to the weighted fusion strategy, where Φ is the state transition matrix, E is the control input matrix, uk is the robot motion control quantity, wk is the filtering process noise, and P... t-n and P t γ represents the robot's historical position sequence, and γ is a fusion coefficient ranging from 0.2 to 0.8.

[0017] In one aspect, the charging system further includes a two-way communication unit, which is used to receive the position signal of the positioning unit, the coordination signal of the computing unit, the scheduling signal of the scheduling unit, and the linkage signal of the linkage control unit;

[0018] The bidirectional communication unit is also used to transmit the operating status signal of the mobile robot and the environmental perception signal of the Internet of Things device back to the computing unit.

[0019] In one aspect, the computing unit is also configured to obtain link parameters from the bidirectional communication unit and construct a link quality index for each communication link of the bidirectional communication unit based on the link parameters. The link quality index is generated by weighted calculation of normalized received signal strength, normalized signal-to-noise ratio and data packet transmission success rate.

[0020] The computing unit outputs a communication parameter adjustment signal to the bidirectional communication unit according to the link quality index, and the bidirectional communication unit receives the communication parameter adjustment signal and dynamically adjusts its own communication parameters.

[0021] In one aspect, the link quality index is LQI;

[0022] LQI = α1*RSSI + α2*SNR + α3*PSR, where α1, α2 and α3 are preset weighting coefficients, RSSI is the normalized received signal strength, SNR is the normalized signal-to-noise ratio, and PSR is the data packet transmission success rate.

[0023] The output of communication parameter adjustment signals to the bidirectional communication unit based on the link quality index includes:

[0024] When the link quality index is lower than the preset threshold, the spreading factor is adjusted according to the preset formula.

[0025] The default formula is SF new =clamp(SF old +β*(LQI th -LQI), SF7, SF12), where SF new The adjusted spreading factor, SF old LQI is the spreading factor before adjustment, and β is the adjustment coefficient, with a value between 0.1 and 0.5. th The preset thresholds are SF7 (minimum spread factor), SF12 (maximum spread factor), and clamp (function that limits the result to the range of SF7-SF12).

[0026] In one aspect, the charging system further includes a power management unit, which is used to perform power allocation and energy management for the charging main body, the positioning unit, the computing unit, the scheduling unit, and the linkage control unit based on the total power supply and the real-time power output control signals of each unit.

[0027] In one aspect, the positioning unit is also used to output a second pulse synchronization signal, and the arithmetic unit receives the pulse synchronization signal output by the positioning unit and performs drift compensation on the local clock.

[0028] In one aspect, the mobile operation robots include outdoor lawnmowers and yard inspection robots, and the IoT devices include soil sensor nodes, sprinkler systems, yard solenoid valves, and lighting fixtures.

[0029] The beneficial effects of this invention are as follows: The charging unit provides basic docking and power replenishment for mobile robots; the positioning unit calculates and outputs robot position signals, laying the positional foundation for multi-robot collaboration; the computing unit receives the position signals, predicts robot trajectories, and outputs coordination signals, providing a predictive basis for scheduling and linkage control; the scheduling unit generates scheduling signals adapted to multi-robot operations based on these coordination signals, avoiding path conflicts and reducing operational timing issues; the linkage control unit synchronously receives the coordination signals and outputs linkage signals to IoT devices, achieving coordination between robot operations and IoT device operation. This constructs an integrated multi-robot collaborative operation system encompassing positioning, prediction, scheduling, and linkage. From basic operational support, position perception, trajectory prediction, orderly multi-robot scheduling to equipment linkage, the entire multi-robot operation process is optimized, reducing operational delays and equipment coordination issues that affect operational efficiency, thereby effectively improving the overall collaborative operation efficiency of multiple robots. Attached Figure Description

[0030] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0031] Figure 1 This is a functional structure diagram of the multi-robot collaborative charging system of this application. Detailed Implementation

[0032] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0033] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0034] like Figure 1As shown, this application provides a multi-robot collaborative charging system. The charging system includes a charging body 10, a positioning unit 20, a computing unit 30, a scheduling unit 50, and a linkage control unit 60. The charging body 10 serves as the basic hardware carrier, providing a physical foundation for the integrated installation of the positioning unit 20, computing unit 30, scheduling unit 50, and linkage control unit 60. Each functional unit can be installed at a preset installation position on the charging body 10. The computing unit 30 is set as the core processing unit of the system, establishing direct signal connections with the positioning unit 20, scheduling unit 50, and linkage control unit 60. The positioning unit 20 is installed at a high, unobstructed position on the charging body 10 to ensure effective reception of satellite positioning signals. The scheduling unit 50 and linkage control unit 60 are installed adjacent to the computing unit 30 to shorten the signal transmission distance. Each unit uses the charging body 10 as its installation support and the computing unit 30 as the core of signal interaction, forming a compact layout that ensures the independent operation of each unit while improving the efficiency and stability of signal transmission.

[0035] The charging unit 10 is used to provide a docking bay and power supply for mobile robots. As the basic hardware carrier of the multi-robot collaborative charging system, the charging unit 10 has a dedicated docking bay for various mobile robots, which can enable the mobile robots to accurately dock at the designated location. At the same time, the charging unit 10 is equipped with a corresponding power supply structure, which can replenish the power of the robot after it docks, ensuring the continuous operation capability of the mobile robots and meeting the needs of multiple robots to charge alternately and operate continuously.

[0036] The positioning unit 20 receives satellite positioning signals and calculates and outputs the position signal of the mobile robot. The positioning unit 20 has the capability to receive and calculate satellite positioning signals, can capture positioning signals transmitted by satellites in real time, and the calculation algorithm processes and analyzes the received signals to obtain the real-time position information of the mobile robot, converting this position information into a standardized position signal for output. The positioning unit 20 can also be understood as an RTK (Real-Time Kinematic) differential base station. The RTK differential base station is the core fixed base station equipment used in RTK real-time dynamic differential positioning technology. It is a hardware component of the positioning unit in this multi-robot collaborative charging system, providing differential correction data for the high-precision position calculation of the positioning unit. It is a key hardware component ensuring high-precision outdoor positioning of the mobile robot, suitable for centimeter / sub-meter level positioning in courtyard scenarios.

[0037] Commonly used algorithms for solving robot positions include least squares, Kalman filtering, extended Kalman filtering, and unscented Kalman filtering. Taking the fusion of satellite positioning and inertial sensing data as an example, least squares substitutes observation data from multiple satellites into a preset equation and finds the parameter value that minimizes the sum of squared errors to obtain the robot's position solution. Kalman filtering first predicts the current position based on the robot's previous position and motion state, and then combines the current satellite and inertial sensing observation data to correct the predicted value through an iterative process of state update and observation update, thus obtaining an accurate real-time position. Extended Kalman filtering and unscented Kalman filtering can also adapt to nonlinear motion and solution models. They first linearize the nonlinear system or fit a probability distribution through sampling points, and then fuse multi-source data according to filtering iteration logic to complete the robot's position solution. All these algorithms solve the position information of the mobile robot by performing mathematical modeling and iterative calculations on the received positioning signals and sensor data to eliminate data errors.

[0038] The computing unit 30 is used to receive the position signal output by the positioning unit 20, predict the trajectory of the mobile robot, and output coordination signals to the corresponding mobile robot to coordinate multi-robot collaborative operations. As the core processing unit of the system, the computing unit 30 can receive the position signal of the mobile robot output by the positioning unit 20 in real time. Based on the built-in trajectory prediction algorithm, it predicts the subsequent trajectory of the mobile robot according to the robot's real-time position and motion pattern. At the same time, the computing unit 30 generates corresponding coordination signals and sends them to each mobile robot according to the trajectory prediction results and the requirements of multi-robot collaborative operations, so as to coordinate the working status of multiple robots and avoid various conflict problems during the operation.

[0039] Commonly used trajectory prediction algorithms include Kalman filtering, long short-term memory (LSTM) network algorithms, particle filtering, and polynomial fitting. Taking trajectory prediction for multi-robot collaborative operations as an example, the Kalman filtering algorithm first predicts the trajectory state at the current moment based on the robot's position, speed, and other state parameters from the previous moment using state equations. Then, it combines real-time collected position observation data and uses observation equations to correct the predicted value. Through an iterative process of prediction and updating, the robot's short-term accurate trajectory is obtained. The LTM network algorithm first trains and learns from the robot's historical position, trajectory, and other time-series data to discover the robot's motion patterns. Then, based on the trained model, it predicts and outputs the robot's future long-term trajectory. Particle filtering algorithms simulate the robot's trajectory motion state using a large number of random particles. They update and resample the particle weights based on observation data, retaining high-weight particles to fit the robot's actual motion trajectory, thus adapting to nonlinear and non-Gaussian motion scenarios. Polynomial fitting algorithms, on the other hand, substitute the robot's historical position data into a preset polynomial equation, solve the equation coefficients through fitting, and then use the fitted polynomial equation to predict the robot's subsequent motion trajectory. All types of trajectory prediction algorithms rely on the robot's historical motion data and real-time state information, and predict the robot's future motion trajectory through mathematical modeling, model training, or iterative correction.

[0040] The scheduling unit 50 is used to receive the coordination signal from the computing unit 30 and generate the scheduling signal for multi-machine collaborative operation. The scheduling unit 50 establishes a signal transmission connection with the computing unit 30 and can receive the coordination signal output by the computing unit 30 in a timely manner. Combined with the actual working status of multiple mobile robots, and according to the overall requirements of collaborative operation, it rationally plans the working sequence, working path and working time of each robot, and generates a standardized multi-machine collaborative operation scheduling signal, so that each mobile robot can carry out the operation in an orderly manner according to the scheduling requirements.

[0041] The linkage control unit 60 receives coordination signals from the computing unit 30 and outputs linkage signals to the corresponding IoT devices. The linkage control unit 60 can receive coordination signals output by the computing unit 30 in real time. These coordination signals include information such as the mobile robot's position, trajectory, and work plan. Based on this information and the operating steps of the IoT devices, the linkage control unit 60 generates corresponding linkage signals and transmits them to various IoT devices in the courtyard scene. This ensures that the operating status of the IoT devices matches the operating status of the mobile robot, achieving coordinated operation between the robot and the IoT devices.

[0042] In this embodiment, the charging unit 10 provides basic docking and power supply for the mobile robot. The positioning unit 20 calculates and outputs the robot's position signal, laying the positional foundation for multi-robot collaboration. The computing unit 30 receives the position signal, performs robot trajectory prediction, and outputs coordination signals, providing a predictive basis for scheduling and linkage control. The scheduling unit 50 generates scheduling signals adapted to multi-robot operations based on the coordination signals, avoiding conflicts in multi-robot operation paths and reducing problems such as chaotic operation sequences. The linkage control unit 60 synchronously receives the coordination signals and outputs linkage signals to IoT devices, realizing the coordination between robot operations and IoT device operation. This constructs an integrated multi-robot collaborative operation system that combines positioning, prediction, scheduling, and linkage. It optimizes the entire multi-robot operation process from basic operation support, position perception, trajectory prediction, orderly multi-robot scheduling to equipment linkage and coordination, reducing problems such as operation delays and equipment coordination failures that affect operation efficiency, thereby effectively improving the overall collaborative operation efficiency of multi-robots.

[0043] In one embodiment of this application, the positioning unit 20 fuses satellite observation data and inertial sensing data for differential calculation, outputting the position signal of the mobile robot. Different satellite observations are weighted by signal-to-noise ratio, and the observation residuals are corrected according to the weights before outputting the position signal. When calculating the position signal of the mobile robot, the positioning unit 20 does not rely on a single data source, but rather fuses satellite observation data and inertial sensing data. Satellite observation data can obtain the robot's absolute position information, while inertial sensing data can record the robot's motion posture and trajectory changes. The positioning unit 20 combines these two types of data and performs differential calculation, ensuring that the calculated position information simultaneously possesses the reference value of absolute position and the continuity of the motion process, thus improving the overall accuracy of the position calculation.

[0044] The positioning unit 20 processes the fused satellite observation data and inertial sensing data through differential calculation to progressively calculate and determine the real-time position of the mobile robot. After the calculation is completed, the positioning unit 20 converts the position information into a standardized electrical signal, forming a position signal that can be recognized and received by other units in the system, and then outputs the position signal outward.

[0045] When receiving observation data from multiple satellites, the positioning unit 20 performs differentiated processing on the observations from different satellites, specifically using the signal-to-noise ratio (SNR) of each satellite signal as the basis for weight allocation. The SNR of a satellite signal reflects the transmission quality of the signal; satellite observation data with a higher SNR are more accurate, and the positioning unit 20 will assign it a higher weight; satellite observation data with a lower SNR are less reliable, and the positioning unit 20 will assign it a lower weight.

[0046] After assigning weights to the observations from each satellite, the positioning unit 20 corrects the observation residuals generated during the position calculation process based on the corresponding weights. The observation residual is the deviation between the calculated result and the actual position. The positioning unit 20 weights the residuals from different satellite observations according to their weights, reducing the impact of the residuals corresponding to low signal-to-noise ratio observations on the final result, and making the corrected position calculation result more closely match the robot's actual position.

[0047] After weighted correction of the observation residuals, positioning unit 20 generates the final position signal based on the corrected solution. The position signal eliminates some of the errors caused by differences in satellite signal quality, further improving data accuracy and stability. Positioning unit 20 outputs this position signal to ensure that subsequent operations such as trajectory prediction and collaborative scheduling based on position information can be carried out in an orderly and accurate manner.

[0048] In one embodiment of this application, the coordination signal output by the computing unit 30 includes operation cost optimization parameters and trajectory prediction parameters. The coordination signal output by the computing unit 30 integrates two types of core parameters: operation cost optimization parameters and trajectory prediction parameters. The operation cost optimization parameters are used to coordinate the overall operation cost of multiple mobile robots, providing a quantitative basis for cost-level calculations in the scheduling and planning of multi-robot collaborative operations. The trajectory prediction parameters are used to reflect the future movement trajectory trends of each mobile robot, providing a trajectory-level reference for multi-robot operation path planning, conflict avoidance, and IoT device linkage. These two types of parameters work together to provide comprehensive and accurate parameter support for the scheduling and linkage control of multi-robot collaborative operations from both cost and trajectory dimensions.

[0049] The objective function for optimizing the activity cost is J = Σdi + λ1*ΣPconflict(i,j) + λ2*ΣDelayi, and the trajectory prediction parameter is Tpred = γ*T. L +(1-γ)*T K The weighted fusion strategy is defined as follows: di is the remaining path length of mobile robot i, Pconflict(i,j) is the path conflict probability between mobile robots i and j, Delayi is the operation time delay of mobile robot i, λ1 and λ2 are preset weight coefficients, λ1 > λ2, and T... L T represents the long-term trajectory prediction result of the neural network model. K The result is the short-term trajectory prediction of the filtering algorithm, where γ is the fusion coefficient and i and j are the number of robots.

[0050] In one embodiment of this application, when the computing unit 30 predicts the trajectory of the mobile robot, it obtains the short-term trajectory prediction value T through the Kalman filter algorithm. K =Φ*T K-1+E*uk+wk; The long-term trajectory prediction value T is obtained through a long short-term memory network model. L =LSTM(P t-n P t-n+1 , ..., Pt); The fused trajectory prediction results are output to the coordination signal according to the weighted fusion strategy, where Φ is the state transition matrix, E is the control input matrix, uk is the robot motion control quantity, wk is the filtering process noise, P t-n and P t γ represents the robot's historical position sequence, and γ is a fusion coefficient ranging from 0.2 to 0.8.

[0051] In one embodiment of this application, the charging system further includes a bidirectional communication unit 40. The bidirectional communication unit 40 receives position signals from the positioning unit 20, coordination signals from the computing unit 30, scheduling signals from the scheduling unit 50, and linkage signals from the linkage control unit 60. The addition of the bidirectional communication unit 40 to the charging system serves as a signal transmission carrier within the system, enabling directional reception of various signals. It can accurately receive the position signals output by the positioning unit 20, the coordination signals generated by the computing unit 30 based on the position signals, and the scheduling signals planned by the scheduling unit 50 based on the coordination signals. It can also receive linkage signals issued by the linkage control unit 60 to IoT devices. The bidirectional communication unit 40 can be understood as a LoRa (Long Range Radio) gateway. A LoRa gateway is a core networking and data forwarding hardware device based on LoRa low-power wide-area wireless communication technology. It is the hardware module of the bidirectional communication unit in this multi-robot collaborative charging system, specifically adapted to short-distance, low-power, multi-node wireless communication scenarios such as outdoor courtyards. It is a key carrier for realizing bidirectional signal transmission between various functional units, mobile robots, and IoT devices within the system.

[0052] The bidirectional communication unit 40 is also used to transmit the operating status signals of the mobile operation robot and the environmental perception signals of the IoT devices back to the computing unit 30. The bidirectional communication unit 40 has the ability to transmit signals in reverse, to acquire the operating status signals of each mobile operation robot, which reflect the real-time operating status of the robot; it can also collect the environmental perception signals of various IoT devices, which reflect the real-time perception data of the devices. The bidirectional communication unit 40 transmits both types of signals back in a timely manner, providing a real-time data source for the subsequent calculations and signal output of the computing unit 30.

[0053] In one embodiment of this application, the processing unit 30 is further configured to obtain link parameters from the bidirectional communication unit 40, and construct a link quality index for each communication link of the bidirectional communication unit 40 based on the link parameters. The link quality index is generated by weighted calculation of normalized received signal strength, normalized signal-to-noise ratio, and data packet transmission success rate. The processing unit 30 has dual capabilities of signal acquisition and data processing, and can also extract link parameters of each communication link from the bidirectional communication unit 40. The link parameters reflect the basic data of the communication link transmission status. By acquiring the link parameters, the processing unit 30 can grasp the real-time transmission status of each communication link, such as obtaining specific parameters like the signal reception status and data transmission success status of a certain link.

[0054] After obtaining the link parameters of each communication link, the processing unit 30 constructs a link quality index for each communication link of the bidirectional communication unit 40. The link quality index is calculated by weighting three indicators: normalized received signal strength, normalized signal-to-noise ratio, and data packet transmission success rate. The processing unit 30 assigns corresponding weights to the three indicators based on their importance in the communication link quality assessment, and then obtains a link quality index that quantitatively reflects the real-time quality of each communication link through weighted calculation.

[0055] The processing unit 30 outputs a communication parameter adjustment signal to the bidirectional communication unit 40 based on the link quality index. The bidirectional communication unit 40 receives the communication parameter adjustment signal and dynamically adjusts its own communication parameters. The processing unit 30 uses the constructed link quality index as a criterion to analyze whether the transmission quality of each communication link meets the system signal transmission requirements. When an abnormality is detected in the link quality index, failing to meet the requirements for stable signal transmission, the processing unit 30 generates a corresponding communication parameter adjustment signal and outputs it to the bidirectional communication unit 40, providing explicit instructions for parameter adjustment by the bidirectional communication unit 40.

[0056] The bidirectional communication unit 40 can receive communication parameter adjustment signals output by the arithmetic unit 30 in real time. These signals contain specific parameter adjustment requirements and directions. Upon receiving the signal, the bidirectional communication unit 40 dynamically adjusts its own communication parameters according to the adjustment instructions within the signal. This optimizes the transmission status of the communication links, ensuring stable transmission across all links and guaranteeing efficient transmission of various signals within the system.

[0057] In one embodiment of this application, the link quality index is LQI; the link quality index LQI reflects the transmission status of the communication link.

[0058] LQI = α1*RSSI + α2*SNR + α3*PSR, where α1, α2, and α3 are preset weighting coefficients, RSSI is the normalized received signal strength, SNR is the normalized signal-to-noise ratio, and PSR is the data packet transmission success rate. The preset weighting coefficients are proportional coefficients assigned to the three indicators (normalized received signal strength, normalized signal-to-noise ratio, and data packet transmission success rate) in the weighted calculation of the link quality index, reflecting the degree of influence of each indicator on the communication link quality. The preset weighting coefficients can be pre-set according to the actual communication usage scenario. Core indicators affecting link quality will be assigned higher weighting coefficients, while secondary indicators will be assigned relatively lower weighting coefficients. This reasonable allocation ensures that the weighted calculation of the link quality index more closely reflects the actual communication link transmission status, guaranteeing the accuracy of the link quality assessment.

[0059] RSSI, or Normalized Received Signal Strength, is an index obtained by normalizing the original signal strength actually received by the bidirectional communication unit 40. Its value is mapped to a fixed range, eliminating the dimensional differences of the original signal strength values ​​under different communication scenarios. The RSSI index directly reflects the strength of the received signal in the communication link. The higher the normalized value, the stronger the signal received by the bidirectional communication unit 40, and the more stable the signal transmission foundation of the communication link.

[0060] SNR, or Normalized Signal-to-Noise Ratio, is an indicator obtained by normalizing the ratio of effective signal strength to noise signal strength in a communication link. It standardizes the original SNR values ​​for different scenarios, facilitating unified link quality assessment. RSSI reflects the purity of the effective signal in the communication link; a higher normalized value indicates less noise interference with the effective signal, and higher accuracy of the signal received by the bidirectional communication unit 40.

[0061] PSR, or Packet Success Rate, refers to the ratio of the number of data packets successfully transmitted by a bidirectional communication unit 40 within a specified time to the total number of data packets sent. It reflects the actual data transmission efficiency and stability of the communication link. The PSR indicator closely matches the actual usage requirements of the communication link. The higher the value, the lower the data packet loss rate in the link, and the more guaranteed the integrity and timeliness of the transmission of various signals, directly affecting the effectiveness of signal interaction within the system.

[0062] The communication parameter adjustment signal output to the bidirectional communication unit 40 based on the link quality index includes:

[0063] When the link quality index is lower than the preset threshold, the spreading factor is adjusted according to the preset formula.

[0064] The default formula is SF new =clamp(SF old +β*(LQI th-LQI), SF7, SF12), where SF new The adjusted spreading factor, SF old LQI is the spreading factor before adjustment, and β is the adjustment coefficient, with a value between 0.1 and 0.5. th The preset thresholds are SF7 (minimum spread factor) and SF12 (maximum spread factor). `clamp` is a function that limits the result to the range of SF7-SF12. The `clamp` function is a clamping function that limits the calculated value to between the set minimum and maximum values. If the calculated value is within the range, the original value is used; if it exceeds the range, the nearest boundary value is used.

[0065] In one embodiment of this application, the charging system further includes a power management unit. The power management unit is used to perform power allocation and energy control for the charging main body 10, positioning unit 20, computing unit 30, scheduling unit 50, and linkage control unit 60 based on the total power supply and the real-time power output control signals of each unit. The power management unit coordinates the power allocation and energy management of the entire charging system, providing energy assurance for the stable operation of each functional unit within the charging system and preventing single-unit power overload or insufficient power supply.

[0066] The power management unit (Power Management Unit) collects real-time data on the total power supply and the real-time power of each functional unit. The total power supply reflects the system's total available power capacity, while the real-time power of each unit reflects its actual power consumption during operation. Based on the real-time changes in these two types of data, the Power Management Unit generates corresponding power control signals to adapt power allocation to the system's actual power supply and consumption needs. For example, when the total power supply decreases, the Power Management Unit will promptly adjust the control signals to match the overall power supply capacity.

[0067] The power management unit uses the generated power control signals as the basis for energy management. According to the allocation requirements in the signals, it performs power allocation and energy management for various functional units within the charging system. Based on the operating priority and real-time power demand of each unit, the power management unit rationally allocates power supply to ensure that each unit obtains energy as needed.

[0068] In one embodiment of this application, the positioning unit 20 is further configured to output a second pulse synchronization signal, and the arithmetic unit 30 receives the pulse synchronization signal output by the positioning unit 20 and performs drift compensation on the local clock. In addition to calculating and outputting the position signal, the positioning unit 20 also has a synchronization signal output function, capable of outputting a second pulse synchronization signal. The second pulse synchronization signal provides a unified time reference for the system, enabling the time of each functional unit within the system to remain synchronized, ensuring time consistency among units during signal interaction and instruction execution.

[0069] The arithmetic unit 30 establishes a signal connection with the positioning unit 20, and can accurately receive the second pulse synchronization signal output by the positioning unit 20. The arithmetic unit 30 uses the second pulse synchronization signal as the basis for time calibration, performs drift compensation operation on its own local clock, corrects the time deviation generated during the operation of the local clock, and ensures that the clock of the arithmetic unit 30 is always consistent with the unified time reference of the system, so as to ensure that the trajectory prediction, parameter calculation and other tasks of the arithmetic unit 30 are carried out in an accurate time dimension.

[0070] In one embodiment of this application, the mobile operation robot includes an outdoor lawnmower robot and a yard inspection robot, and the Internet of Things (IoT) devices include soil sensor nodes, sprinkler systems, yard solenoid valves, and lighting fixtures. The mobile operation robots adapted to this system are mainly outdoor lawnmower robots and yard inspection robots. The IoT devices linked to the system include soil sensor nodes, sprinkler systems, yard solenoid valves, and lighting fixtures. These various IoT devices can adjust their operating status according to the system's linkage signals to achieve collaborative operation with the mobile operation robots.

[0071] In addition, the positioning unit 20 in this application is equipped with a real-time signal quality detection module. This module continuously monitors the signal-to-noise ratio (SNR) of the satellite signal. When the SNR is detected to be lower than a preset value, indicating that the satellite signal is blocked or lost, the positioning unit 20 automatically increases the weight of inertial sensor data in the differential calculation, maintaining the continuity of position calculation based on the inertial sensor data. When the SNR recovers to above the preset value, the positioning unit 20 dynamically reduces the calculation weight of the inertial sensor data, reverting to the original fusion calculation strategy of satellite observation data and inertial sensor data. Simultaneously, the positioning unit 20, considering the terrain characteristics of the courtyard operation scenario, presets inertial sensor drift correction coefficients for different terrains such as flat roads and slopes, performing terrain adaptation correction on the motion data collected by the inertial sensor. This eliminates the cumulative error of sensor data caused by terrain differences, further improving the accuracy of position calculation.

[0072] When generating the coordination signal, the computing unit 30 also incorporates the remaining battery power data, unfinished task percentage data, and task area urgency data of the mobile robots transmitted back by the bidirectional communication unit 40 into the calculation scope, establishing a charging priority quantification model to calculate the charging priority value for each mobile robot. The weight of the remaining battery power data is higher than that of the task area urgency and unfinished task percentage. After receiving the coordination signal containing the charging priority parameters, the scheduling unit 50 assigns a charging order to multiple mobile robots according to the charging priority value. Simultaneously, based on the task plan of each robot, it determines the charging time for a single robot and the switching point between task and charging, prioritizing charging for robots with low battery power and high urgency tasks to avoid robot shutdowns due to insufficient power during operation and to reduce delays in high-urgency tasks.

[0073] The processing unit 30 performs a health rating on all communication links of the bidirectional communication unit 40 based on three indicators: link quality index, data transmission delay, and packet loss rate, classifying them into three rating levels: excellent, qualified, and faulty. When a communication link is detected to be faulty, the processing unit 30 outputs a link switching signal to the bidirectional communication unit 40. Upon receiving the signal, the bidirectional communication unit 40 automatically switches the signal transmission task of the faulty link to a redundant backup link, ensuring uninterrupted signal transmission. The bidirectional communication unit 40 can dynamically switch between multiple channels. When the interference signal strength of the current communication channel is detected to be higher than a preset value, it automatically switches to a non-interference or low-interference communication channel according to a preset channel priority order without manual intervention. Simultaneously, when adjusting the spreading factor according to a preset formula, the bidirectional communication unit 40 also performs coordinated adjustments to the communication bandwidth and transmission power. When the spreading factor is increased, the communication bandwidth is reduced and the transmission power is slightly increased; when the spreading factor is decreased, the communication bandwidth is increased and the transmission power is reduced, achieving multi-dimensional optimal matching of communication parameters and improving the stability of link signal transmission.

[0074] While outputting linkage signals to IoT devices, the linkage control unit 60 continuously receives operational status feedback signals from each IoT device. Soil sensor nodes transmit environmental perception feedback signals such as soil moisture and temperature; sprinkler systems and yard solenoid valves transmit on / off operational status feedback signals; and lighting fixtures transmit brightness and on / off operational status feedback signals. The linkage control unit 60 transmits all feedback signals to the computing unit 30 in real time. Based on the operational status feedback signals from the IoT devices, the computing unit 30 dynamically adjusts the operating trajectory and parameters of the mobile operation robot. When it detects that the sprinkler system is on, it immediately adjusts the operating trajectories of the corresponding lawnmower robot and inspection robot to avoid the sprinkler operation area. When it detects that the soil moisture level from the soil sensor node is higher than a preset threshold, it suspends yard inspection operations in that area. When it detects that the brightness of the lighting fixtures is lower than a preset value, it adjusts the night vision operation parameters of the inspection robot, thus achieving closed-loop control of the operation linkage between IoT devices and the mobile operation robot.

[0075] The power management unit monitors the operating load of each functional unit and the charging periods of multiple mobile robots in real time. Based on the monitoring data, it performs load balancing power distribution. When it detects that multiple robots are charging simultaneously on the charging unit 10 and the positioning unit 20 is in a high-power signal receiving state, it appropriately reduces the power supply of the scheduling unit 50 and the linkage control unit 60 to reduce the overall system power overload caused by multiple units operating at high power simultaneously. When the charging system is in a light-load state, i.e., no mobile robots are docked for charging and only the positioning unit 20 is in a low-power signal monitoring state, the power management unit automatically switches the scheduling unit 50 and the linkage control unit 60 to a low-power sleep mode to reduce the overall system energy consumption. When it detects that a robot is about to dock for charging or the positioning unit 20 needs to work at high power, it immediately wakes up the units in sleep mode and restores normal power supply. At the same time, the power management unit identifies the battery type and remaining power of the mobile robots, and presets the power parameters for constant current and constant voltage charging stages for lithium batteries and lead-acid batteries respectively. It automatically adjusts the charging power according to the remaining power of the robot, improving charging efficiency while protecting the lifespan of the robot battery.

[0076] The computing unit 30 is equipped with a trajectory prediction model self-learning module. This module continuously collects actual operational trajectory data and trajectory prediction results from each mobile robot, calculates the deviation between the predicted and actual trajectories, and dynamically iteratively optimizes the state transition matrix and control input matrix parameters of the Kalman filter algorithm based on the deviation data. Simultaneously, it updates and adjusts the training parameters of the Long Short-Term Memory network model, gradually reducing trajectory prediction deviation and improving accuracy. During trajectory prediction, the computing unit 30 also incorporates obstacle avoidance constraints specific to the courtyard operation scenario. The location information of fixed obstacles such as flower beds, fences, and artificial hills within the courtyard is entered into a preset database. When predicting the movement trajectory of the mobile robot, it automatically avoids the fixed obstacle areas in the database, outputting a collision-free predicted trajectory and improving the scheduling efficiency of multi-robot collaborative operations.

[0077] The second pulse synchronization signal output by the positioning unit 20 is transmitted not only to the arithmetic unit 30, but also synchronously to the bidirectional communication unit 40, the scheduling unit 50, and the linkage control unit 60, achieving full coverage of the second pulse synchronization signal across all functional units of the system. The arithmetic unit 30, the bidirectional communication unit 40, the scheduling unit 50, and the linkage control unit 60 all perform drift compensation operations on their local clocks based on the received second pulse synchronization signal. Each unit corrects the time deviation generated during the operation of its local clock according to the time reference of the second pulse synchronization signal, achieving precise time synchronization of all functional units in the entire system. This avoids problems such as signal interaction delays, asynchronous execution of scheduling instructions, and deviations in linkage signal output caused by inconsistent time references among units, ensuring the efficient collaborative operation of all units in the system.

[0078] The bidirectional communication unit 40 transmits the mobile robot's operating status signal, including the robot's fault code, motion status data, battery temperature data, and work execution data. The computing unit 30 monitors this data in real time and presets anomaly judgment thresholds for fault codes, motion status, and battery temperature. When the robot's battery temperature exceeds the preset threshold, or when anomalies such as lag or deviation occur in the motion status, or when a robot fault code is received, the computing unit 30 immediately generates an emergency coordination signal. The emergency coordination signal includes instruction parameters to suspend work and return to the charging bay. After receiving the emergency coordination signal, the scheduling unit 50 outputs an emergency scheduling signal to the abnormal robot, instructing the robot to suspend its current work and return to the charging unit 10's docking bay according to the optimal path. Simultaneously, the linkage control unit 60 outputs an anomaly warning signal to the backend terminal, feeding back the abnormal robot's device number, anomaly type, and current location, realizing integrated processing of mobile robot anomaly detection, emergency scheduling, and warning reporting.

[0079] The multi-robot collaborative charging system of this application has three preset operating modes: daily operation mode, severe weather mode, and holiday low-load mode. It can adaptively switch between these modes based on environmental sensing signals or manually set commands. When the environmental sensing signals from the IoT devices transmitted back by the two-way communication unit 40 show rainfall detected by the rain sensor or wind speed detected by the wind speed sensor exceeding a preset value, it automatically switches to severe weather mode. The positioning unit 20 increases the satellite signal anti-interference level, the two-way communication unit 40 increases the spreading factor and switches to an anti-interference channel, the scheduling unit 50 immediately instructs all outdoor mobile robots to suspend operations and return to their charging positions, and the linkage control unit 60 instructs the sprinkler system and yard solenoid valve to close. When the environmental sensing signals return to normal, it automatically switches back to daily operation mode. When a manually set command for holiday low-load mode is received, the signal receiving frequency of the positioning unit 20 is reduced, the power management unit switches non-core units to low-power mode, and the scheduling unit 50 adjusts the robot's operation plan and charging scheduling strategy according to holiday operation requirements, reducing unnecessary operations and energy consumption to adapt to the low-load operation requirements of holidays. When a manually set command to resume operation is received, it switches back to daily operation mode.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A multi-robot collaborative charging system, characterized in that, The charging system includes: A charging unit, which provides a docking bay and power supply for the mobile robot. The positioning unit is used to receive satellite positioning signals and calculate and output the position signal of the mobile operation robot; The computing unit is used to receive the position signal output by the positioning unit, predict the trajectory of the mobile robot, and output a coordination signal to the corresponding mobile robot to coordinate multi-robot collaborative operation. A scheduling unit is used to receive coordination signals from the computing unit and generate scheduling signals for multi-machine collaborative operations. The linkage control unit is used to receive coordination signals from the computing unit and output linkage signals to the corresponding Internet of Things devices.

2. The charging system according to claim 1, characterized in that, The positioning unit integrates satellite observation data and inertial sensor data to perform differential calculations, outputs the position signal of the mobile operation robot, and weights different satellite observations with signal-to-noise ratio as the weight, corrects the observation residuals according to the weights, and then outputs the position signal.

3. The charging system according to claim 1, characterized in that, The coordination signal output by the computing unit includes job cost optimization parameters and trajectory prediction parameters; The operation cost optimization parameters are defined with the objective function J = Σdi + λ1 * ΣPconflict(i, j) + λ2 * ΣDelayi, and the trajectory prediction parameters are defined with Tpred = γ * T. L +(1-γ)*T K The weighted fusion strategy is defined as follows: di is the remaining path length of mobile robot i, Pconflict(i,j) is the path conflict probability between mobile robots i and j, Delayi is the operation time delay of mobile robot i, λ1 and λ2 are preset weight coefficients, λ1 > λ2, and T... L T represents the long-term trajectory prediction result of the neural network model. K The result is the short-term trajectory prediction of the filtering algorithm, where γ is the fusion coefficient and i and j are the number of robots.

4. The charging system according to claim 3, characterized in that, When the computing unit predicts the trajectory of the mobile robot... The short-term trajectory prediction value T is obtained through the Kalman filter algorithm. K =Φ*T K-1 +E*uk+wk; The long-term trajectory prediction value T is obtained through the long short-term memory network model. L =LSTM(P t−n P t−n+1 ..., Pt); The fused trajectory prediction results are output to the coordination signal according to the weighted fusion strategy, where Φ is the state transition matrix, E is the control input matrix, uk is the robot motion control quantity, wk is the filtering process noise, and P... t-n and P t γ represents the robot's historical position sequence, and γ is a fusion coefficient ranging from 0.2 to 0.

8.

5. The charging system according to claim 1, characterized in that, The charging system also includes a two-way communication unit, which is used to receive the position signal of the positioning unit, the coordination signal of the computing unit, the scheduling signal of the scheduling unit, and the linkage signal of the linkage control unit. The bidirectional communication unit is also used to transmit the operating status signal of the mobile robot and the environmental perception signal of the Internet of Things device back to the computing unit.

6. The charging system according to claim 5, characterized in that, The computing unit is also used to obtain link parameters from the bidirectional communication unit and construct a link quality index for each communication link of the bidirectional communication unit based on the link parameters. The link quality index is generated by weighted calculation of normalized received signal strength, normalized signal-to-noise ratio and data packet transmission success rate. The computing unit outputs a communication parameter adjustment signal to the bidirectional communication unit according to the link quality index, and the bidirectional communication unit receives the communication parameter adjustment signal and dynamically adjusts its own communication parameters.

7. The charging system according to claim 6, characterized in that, The link quality index is LQI, LQI=α1*RSSI+α2*SNR+α3*PSR, where α1, α2 and α3 are preset weighting coefficients, RSSI is the normalized received signal strength, SNR is the normalized signal-to-noise ratio, and PSR is the data packet transmission success rate. The output of communication parameter adjustment signals to the bidirectional communication unit based on the link quality index includes: When the link quality index is lower than the preset threshold, the spreading factor is adjusted according to the preset formula. The default formula is SF new =clamp(SF old +β*(LQI th -LQI), SF7, SF12), where SF new The adjusted spreading factor, SF old LQI is the spreading factor before adjustment, and β is the adjustment coefficient, with a value between 0.1 and 0.

5. th The preset thresholds are SF7 (minimum spread factor), SF12 (maximum spread factor), and clamp (function that limits the result to the range of SF7 to SF12).

8. The charging system according to claim 1, characterized in that, The charging system also includes a power management unit, which is used to perform power allocation and energy management for the charging main body, the positioning unit, the computing unit, the scheduling unit, and the linkage control unit based on the total power supply and the real-time power output control signals of each unit.

9. The charging system according to claim 1, characterized in that, The positioning unit is also used to output a second pulse synchronization signal, and the arithmetic unit receives the pulse synchronization signal output by the positioning unit and performs drift compensation on the local clock.

10. The charging system according to any one of claims 1 to 9, characterized in that, The mobile operation robots include outdoor lawn mowing robots and yard inspection robots, and the Internet of Things devices include soil sensor nodes, sprinkler systems, yard solenoid valves, and lighting fixtures.

Images