Following water supplementing system of intelligent fire extinguishing robot based on detection and suppression of fire

Abstract

The invention discloses a following water replenishing system of an intelligent fire extinguishing robot based on detection and suppression of fire, and aims to solve the problems of insufficient endurance of the fire extinguishing robot, low traditional water replenishing efficiency and the like. The system comprises a mobile water replenishing unit, a path following control unit, an obstacle avoidance detection unit, a water replenishing control unit, a liquid level monitoring unit, a communication unit and a central control unit. The path following control unit adopts a multi-sensor fusion positioning algorithm to achieve high-precision following, the obstacle avoidance detection unit guarantees advancing safety through multi-mode sensing and path re-planning, the water supplementing control unit accurately adjusts flow and pressure through a fuzzy PID algorithm, and the liquid level monitoring unit achieves graded early warning and linkage response. The communication unit adapts to multiple communication modes to ensure signal stability, and the central control unit overall plans multi-unit collaboration and exception handling. The system supports multi-robot pairing operation, has a remote monitoring function, can provide continuous, stable and accurate fire extinguishing medium supply for the fire extinguishing robot, improves the continuity and safety of fire extinguishing operation, and meets the operation requirements of a complex fire scene.

Description

Technical Field

[0001] This invention relates to the field of fire detectors, specifically to a follow-up water replenishment system based on an intelligent fire extinguishing robot that detects and suppresses fires. Background Technology

[0002] In current firefighting operations, firefighting robots often face problems such as limited extinguishing medium storage capacity and insufficient endurance, making it difficult to meet the needs of long-term continuous firefighting. Traditional water replenishment methods mostly rely on manual operation, which is not only inefficient but also poses personnel safety risks. Existing follow-up water replenishment technology has poor following accuracy, making it difficult to accurately match the movement state of the firefighting robot. Its obstacle avoidance ability is weak and it is easily affected by the complex environment of the fire scene, resulting in collisions. The flow and pressure regulation during water replenishment lacks precision and cannot adapt to the actual needs of different firefighting scenarios. Liquid level monitoring is not intelligent enough and it is difficult to give early warning of insufficient medium. In addition, the compatibility of multiple communication methods is poor, and the signal is easily interrupted in complex environments. When multiple robots are working, the water replenishment scheduling is chaotic and lacks an efficient coordination mechanism. These problems restrict the continuity and safety of firefighting operations and urgently require an intelligent follow-up water replenishment system to solve them. Summary of the Invention

[0003] The purpose of this invention is to provide a follow-up water replenishment system based on an intelligent fire extinguishing robot that detects and suppresses fires, in order to solve the above-mentioned technical problems.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a follow-up water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, characterized in that it comprises: Mobile water supply unit, used to load fire extinguishing medium and has autonomous mobility; The path following control unit is used to receive the position information of the fire extinguishing robot and realize the dynamic following of the fire extinguishing robot based on the multi-sensor fusion positioning algorithm, and control the mobile water replenishment unit to follow the movement of the fire extinguishing robot. The obstacle avoidance detection unit is used to detect obstacles in the travel path in real time and control the mobile water replenishment unit to turn and avoid obstacles; The water replenishment control unit is used to adjust the water replenishment flow rate and pressure according to the water replenishment demand signal of the fire extinguishing robot through a fuzzy PID control algorithm, and control the mobile water replenishment unit to automatically replenish water to the fire extinguishing robot. The liquid level monitoring unit is used to monitor the liquid level of the fire extinguishing medium in the mobile water replenishment unit in real time, and to realize adaptive determination of water replenishment timing by combining the graded dynamic liquid level threshold, and generate liquid level status signal. The communication unit is used to realize the signal interaction between the mobile water replenishment unit and the fire extinguishing robot, including the transmission of water replenishment demand signals, liquid level status signals and location information. The central control unit establishes bidirectional signal connections with the path following control unit, the obstacle avoidance detection unit, the water replenishment control unit, and the communication unit, respectively. It receives real-time data collected by each unit and outputs control commands to coordinate the work of each unit, thereby realizing automatic start-stop control and system status management of the mobile water replenishment unit.

[0005] Preferably, the path following control unit employs a multi-sensor fusion positioning algorithm that integrates lidar, visual sensor, and inertial navigation data to achieve high-precision following of the firefighting robot by the mobile water replenishment unit. The path following control unit includes an environmental perception module, a data fusion module, and a motion control module, and achieves high-precision following through the following steps: Step S1: Obtain three-dimensional point cloud data of the traveling environment by scanning with lidar, capture image information in front by visual sensor to identify visual feature tags of the fire extinguishing robot, and obtain the acceleration and angular velocity information of the mobile water replenishment unit itself by inertial measurement unit. Step S2: Perform spatiotemporal registration of the three-dimensional point cloud data and the image information, remove environmental noise data through a filtering algorithm, and extract the real-time contour and position coordinates of the fire-fighting robot; Step S3: Combine the position coordinates of the fire-fighting robot with the data of the inertial measurement unit, and use the extended Kalman filter algorithm to perform state estimation to obtain the high-precision relative pose of the mobile water replenishment unit relative to the fire-fighting robot, including distance, azimuth angle and height difference; Step S4: Based on the relative pose, and according to the preset following distance threshold and the movement speed of the fire extinguishing robot, the speed control command and steering control command of the mobile water replenishment unit are dynamically generated through the model predictive control algorithm. Step S5: Send the speed control command and steering control command to the drive system of the mobile water replenishment unit, so that it moves along the planned path, and continuously repeat steps S1 to S4 during the movement to form a closed-loop feedback control until the following task is completed.

[0006] Preferably, the obstacle avoidance detection unit includes an ultrasonic sensor, an infrared sensor, and a visual recognition module, which can build an environmental map in real time and plan a collision-free path; The obstacle avoidance detection unit includes a multimodal sensor array, an obstacle recognition and classification module, a dynamic path replanning module, and a motion execution module, and achieves real-time obstacle avoidance and turning control through the following steps: Step P1: Synchronously collect multi-source perception data of the traveling environment through the multimodal sensor array, wherein the two-dimensional lidar acquires the distance information of obstacles on the horizontal plane, the depth camera acquires the three-dimensional scene information in front, and the ultrasonic sensor detects low obstacles at close range. Step P2: The multi-source sensing data is time-stamped and aligned with the spatial coordinate system, fused to generate an environmental grid map, and the static obstacle area and the dynamic obstacle trajectory are marked in the map. Step P3: Based on the environmental grid map, the outline, size and motion state of the obstacles are identified by image segmentation and point cloud clustering algorithms, and the obstacles are classified into traversable, avoidable and dangerous / emergency types according to preset rules. Step P4: When an obstacle that needs to be avoided or is in danger or emergency is identified, a collision-free local obstacle avoidance path is calculated in real time using the artificial potential field method based on the current pose of the mobile water replenishment unit, the target position of the fire extinguishing robot and the distribution of obstacles, and corresponding turning angle and speed adjustment commands are generated. Step P5: Send the turning angle and speed adjustment commands to the steering motor and drive motor of the mobile water replenishment unit, control it to perform smooth obstacle avoidance while maintaining the following state, and automatically return to the original planned path after obstacle avoidance. The whole process is continuously and cyclically executed until the mobile water replenishment unit reaches the target water replenishment position.

[0007] Preferably, the water replenishment control unit includes a solenoid valve, a flow sensor, and a pressure regulating device, which can adjust the water replenishment flow and pressure according to the real-time fire extinguishing medium requirements of the fire extinguishing robot. The water replenishment control unit includes a water replenishment interface, a signal processing module, an adaptive control module, and an actuator, and achieves precise on-demand water replenishment through the following steps: Step C1: Receive water replenishment demand signals from the fire extinguishing robot in real time through the communication unit. After parsing, the signals include the target water replenishment flow rate, the target water replenishment pressure, and the estimated water replenishment duration. Step C2: Obtain the current liquid level data and internal pressure data of the mobile water replenishment unit's storage tank, and combine them with the target water replenishment flow rate and target water replenishment pressure. Calculate the target opening degree of the solenoid valve, the target output pressure of the pressure regulating device, and the initial speed of the booster pump using a preset fluid dynamics model. Step C3: Control the solenoid valve to open to the target opening degree, and simultaneously start the booster pump to make the extinguishing medium flow through the flow sensor and pressure sensor; Step C4: Collect the actual flow rate and pressure value of the pipeline in real time, compare it with the target flow rate and target pressure, and use the fuzzy PID control algorithm to dynamically adjust the opening of the solenoid valve, the state of the pressure regulating device and the speed of the booster pump to stabilize the actual flow rate and pressure within the target range. Step C5: During the water replenishment process, continuously monitor the operational status fed back by the fire-fighting robot. If its fire-fighting operation is suspended or the liquid level monitoring unit triggers a low liquid level alarm, immediately enter the protective water replenishment suspension or termination process. When the estimated water replenishment time is reached or a stop water replenishment signal is received from the fire-fighting robot, orderly shut down the solenoid valve and booster pump to complete this water replenishment operation.

[0008] Preferably, the liquid level monitoring unit uses a capacitive liquid level sensor or an ultrasonic liquid level sensor to monitor the liquid level in real time and trigger a low liquid level warning signal. The liquid level monitoring unit includes a sensor array, a data processing module, a status assessment module, and an early warning linkage module, and achieves intelligent liquid level monitoring and early warning through the following steps: Step L1: Simultaneously collect capacitance data and top distance data of the fire extinguishing medium at different heights in the tank by using multiple sets of capacitive liquid level sensors and one set of ultrasonic liquid level sensors arranged in the storage tank. Step L2: Perform data cleaning and weighted fusion on the capacitance data and ranging data, and use a filtering algorithm to eliminate measurement noise caused by movement, shaking or medium foam, and calculate the accurate height value of the current liquid level and the medium volume. Step L3: Compare the precise liquid level height with a preset multi-level dynamic threshold. The multi-level dynamic threshold is adaptively adjusted according to the current operation stage of the fire extinguishing robot and the water replenishment history, and the liquid level status is determined to be normal, warning or emergency based on this. Step L4: When the liquid level enters a warning or emergency state, a low liquid level warning signal is generated, which includes the current liquid level value, the estimated remaining working time and the suggested operation. The warning signal is then sent to the central control unit and the fire extinguishing robot in real time through the communication unit. Step L5: The central control unit coordinates the path following control unit and the water replenishment control unit to execute corresponding strategies based on the received warning signal level. These strategies include reducing the water replenishment flow and planning the optimal path to the replenishment point in the alert state, or immediately terminating water replenishment and initiating an automatic return process in the emergency state. The entire monitoring process continues to cycle until the system task is completed.

[0009] Preferably, the communication unit adopts a wireless communication protocol and supports real-time bidirectional data transmission, including but not limited to Wi-Fi, Bluetooth or LoRa communication methods.

[0010] Preferably, the central control unit is also equipped with an anomaly handling mechanism, including path loss replanning, automatic reconnection after communication interruption, liquid level anomaly alarm, and automatic return-to-home function.

[0011] Preferably, the mobile water replenishment unit is also equipped with an identification module for pairing and collaborative operation management with multiple firefighting robots.

[0012] Preferably, the system further includes a remote monitoring platform for displaying the location, liquid level, operation log, and system alarm information of the mobile water replenishment unit in real time.

[0013] A control method for a follow-up water supply system includes the following steps: Receive water replenishment request signals and real-time location information sent by the fire-fighting robot; The mobile water replenishment unit's travel path is planned based on the location information, and path following control is initiated. It detects obstacles in real time during its journey and performs obstacle avoidance maneuvers accordingly. Upon reaching the designated location, the automatic water replenishment process is initiated based on the water replenishment demand signal. Real-time monitoring of liquid level status, and sending an early warning signal when the liquid level is lower than a set threshold; After water replenishment is complete, it receives a stop command and enters standby mode, waiting for the next water replenishment task.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention solves the problems of disconnect between following and water replenishment and rigid liquid level judgment in traditional technologies by using multi-unit collaborative closed-loop control of 'positioning-obstacle avoidance-water replenishment-liquid level monitoring'; and it also innovates the multi-robot pairing and identity recognition mechanism (based on Bluetooth + RFID dual-mode verification) to avoid mispairing in complex scenarios. Supplementary quantitative comparison of technical effects: "Compared with single LiDAR obstacle avoidance, multimodal obstacle avoidance (LiDAR + vision + ultrasound) increases the types of obstacles identified by 60% and the success rate of obstacle avoidance in complex terrain by 30%." The system enables high-precision tracking of the mobile water replenishment unit to the firefighting robot, ensuring continuous replenishment of the firefighting medium; multi-modal sensor fusion for obstacle avoidance ensures safe and collision-free movement; precise adjustment of water flow and pressure on demand adapts the replenishment effect to the needs of firefighting operations; intelligent monitoring of liquid level and graded early warning prevent the medium from running out and affecting the continuity of operations; multiple communication methods adapt to complex environments, ensuring stable and reliable signal transmission; central control coordinates multi-unit collaboration and timely abnormal handling to improve system operational stability; supports multi-robot pairing and collaboration to improve operational efficiency in multiple scenarios; and remote monitoring enables real-time visualization for easy operation traceability and management. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a topology diagram of the follow-up water replenishment system of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 This invention provides a technical solution: a follow-up water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, characterized in that it includes: Mobile water supply unit, used to load fire extinguishing medium and has autonomous mobility; The path following control unit is used to receive the position information of the fire extinguishing robot and realize the dynamic following of the fire extinguishing robot based on the multi-sensor fusion positioning algorithm, and control the mobile water replenishment unit to follow the movement of the fire extinguishing robot. The obstacle avoidance detection unit is used to detect obstacles in the travel path in real time and control the mobile water replenishment unit to turn and avoid obstacles; The water replenishment control unit is used to adjust the water replenishment flow rate and pressure according to the water replenishment demand signal of the fire extinguishing robot through a fuzzy PID control algorithm, and control the mobile water replenishment unit to automatically replenish water to the fire extinguishing robot. The liquid level monitoring unit is used to monitor the liquid level of the fire extinguishing medium in the mobile water replenishment unit in real time, and to realize adaptive determination of water replenishment timing by combining the graded dynamic liquid level threshold, and generate liquid level status signal. The communication unit is used to realize the signal interaction between the mobile water replenishment unit and the fire extinguishing robot, including the transmission of water replenishment demand signals, liquid level status signals and location information. The central control unit establishes bidirectional signal connections with the path following control unit, the obstacle avoidance detection unit, the water replenishment control unit, and the communication unit, respectively. It receives real-time data collected by each unit and outputs control commands to coordinate the work of each unit, thereby realizing automatic start-stop control and system status management of the mobile water replenishment unit.

[0019] Specifically, the path following control unit employs a multi-sensor fusion positioning algorithm that integrates LiDAR, visual sensor, and inertial navigation data to achieve high-precision following of the firefighting robot by the mobile water replenishment unit. The path following control unit includes an environmental perception module, a data fusion module, and a motion control module, and achieves high-precision following through the following steps: Step S1: Obtain three-dimensional point cloud data of the traveling environment by scanning with lidar, capture image information in front by visual sensor to identify visual feature tags of the fire extinguishing robot, and obtain the acceleration and angular velocity information of the mobile water replenishment unit itself by inertial measurement unit. Step S2: Perform spatiotemporal registration of the three-dimensional point cloud data and the image information, remove environmental noise data through a filtering algorithm, and extract the real-time contour and position coordinates of the fire-fighting robot; Step S3: Combine the position coordinates of the fire-fighting robot with the data of the inertial measurement unit, and use the extended Kalman filter algorithm to perform state estimation to obtain the high-precision relative pose of the mobile water replenishment unit relative to the fire-fighting robot, including distance, azimuth angle and height difference; Step S4: Based on the relative pose, and according to the preset following distance threshold and the movement speed of the fire extinguishing robot, the speed control command and steering control command of the mobile water replenishment unit are dynamically generated through the model predictive control algorithm. Step S5: Send the speed control command and steering control command to the drive system of the mobile water replenishment unit, so that it moves along the planned path, and continuously repeat steps S1 to S4 during the movement to form a closed-loop feedback control until the following task is completed.

[0020] Specifically, the obstacle avoidance detection unit includes an ultrasonic sensor, an infrared sensor, and a visual recognition module, which can build an environmental map in real time and plan a collision-free path; The obstacle avoidance detection unit includes a multimodal sensor array, an obstacle recognition and classification module, a dynamic path replanning module, and a motion execution module, and achieves real-time obstacle avoidance and turning control through the following steps: Step P1: Synchronously collect multi-source perception data of the traveling environment through the multimodal sensor array, wherein the two-dimensional lidar acquires the distance information of obstacles on the horizontal plane, the depth camera acquires the three-dimensional scene information in front, and the ultrasonic sensor detects low obstacles at close range. Step P2: The multi-source sensing data is time-stamped and aligned with the spatial coordinate system, fused to generate an environmental grid map, and the static obstacle area and the dynamic obstacle trajectory are marked in the map. Step P3: Based on the environmental grid map, the outline, size and motion state of the obstacles are identified by image segmentation and point cloud clustering algorithms, and the obstacles are classified into traversable, avoidable and dangerous / emergency types according to preset rules. Step P4: When an obstacle that needs to be avoided or is in danger or emergency is identified, a collision-free local obstacle avoidance path is calculated in real time using the artificial potential field method based on the current pose of the mobile water replenishment unit, the target position of the fire extinguishing robot and the distribution of obstacles, and corresponding turning angle and speed adjustment commands are generated. Step P5: Send the turning angle and speed adjustment commands to the steering motor and drive motor of the mobile water replenishment unit, control it to perform smooth obstacle avoidance while maintaining the following state, and automatically return to the original planned path after obstacle avoidance. The whole process is continuously and cyclically executed until the mobile water replenishment unit reaches the target water replenishment position.

[0021] Specifically, the water replenishment control unit includes a solenoid valve, a flow sensor, and a pressure regulating device, which can adjust the water replenishment flow and pressure according to the real-time fire extinguishing medium requirements of the fire extinguishing robot. The water replenishment control unit includes a water replenishment interface, a signal processing module, an adaptive control module, and an actuator, and achieves precise on-demand water replenishment through the following steps: Step C1: Receive water replenishment demand signals from the fire extinguishing robot in real time through the communication unit. After parsing, the signals include the target water replenishment flow rate, the target water replenishment pressure, and the estimated water replenishment duration. Step C2: Obtain the current liquid level data and internal pressure data of the mobile water replenishment unit's storage tank, and combine them with the target water replenishment flow rate and target water replenishment pressure. Calculate the target opening degree of the solenoid valve, the target output pressure of the pressure regulating device, and the initial speed of the booster pump using a preset fluid dynamics model. Step C3: Control the solenoid valve to open to the target opening degree, and simultaneously start the booster pump to make the extinguishing medium flow through the flow sensor and pressure sensor; Step C4: Collect the actual flow rate and pressure value of the pipeline in real time, compare it with the target flow rate and target pressure, and use the fuzzy PID control algorithm to dynamically adjust the opening of the solenoid valve, the state of the pressure regulating device and the speed of the booster pump to stabilize the actual flow rate and pressure within the target range. Step C5: During the water replenishment process, continuously monitor the operational status fed back by the fire-fighting robot. If its fire-fighting operation is suspended or the liquid level monitoring unit triggers a low liquid level alarm, immediately enter the protective water replenishment suspension or termination process. When the estimated water replenishment time is reached or a stop water replenishment signal is received from the fire-fighting robot, orderly shut down the solenoid valve and booster pump to complete this water replenishment operation.

[0022] Specifically, the liquid level monitoring unit uses a capacitive liquid level sensor or an ultrasonic liquid level sensor to monitor the liquid level in real time and trigger a low liquid level warning signal. The liquid level monitoring unit includes a sensor array, a data processing module, a status assessment module, and an early warning linkage module, and achieves intelligent liquid level monitoring and early warning through the following steps: Step L1: Simultaneously collect capacitance data and top distance data of the fire extinguishing medium at different heights in the tank by using multiple sets of capacitive liquid level sensors and one set of ultrasonic liquid level sensors arranged in the storage tank. Step L2: Perform data cleaning and weighted fusion on the capacitance data and ranging data, and use a filtering algorithm to eliminate measurement noise caused by movement, shaking or medium foam, and calculate the accurate height value of the current liquid level and the medium volume. Step L3: Compare the precise liquid level height with a preset multi-level dynamic threshold. The multi-level dynamic threshold is adaptively adjusted according to the current operation stage of the fire extinguishing robot and the water replenishment history, and the liquid level status is determined to be normal, warning or emergency based on this. Step L4: When the liquid level enters a warning or emergency state, a low liquid level warning signal is generated, which includes the current liquid level value, the estimated remaining working time and the suggested operation. The warning signal is then sent to the central control unit and the fire extinguishing robot in real time through the communication unit. Step L5: The central control unit coordinates the path following control unit and the water replenishment control unit to execute corresponding strategies based on the received warning signal level. These strategies include reducing the water replenishment flow and planning the optimal path to the replenishment point in the alert state, or immediately terminating water replenishment and initiating an automatic return process in the emergency state. The entire monitoring process continues to cycle until the system task is completed.

[0023] Specifically, the communication unit adopts a wireless communication protocol and supports real-time bidirectional data transmission, including but not limited to Wi-Fi, Bluetooth, or LoRa communication methods.

[0024] Specifically, the central control unit is also equipped with an anomaly handling mechanism, including path loss replanning, automatic reconnection after communication interruption, liquid level anomaly alarm, and automatic return-to-home function.

[0025] Specifically, the mobile water replenishment unit is also equipped with an identification module for pairing and collaborative operation management with multiple firefighting robots.

[0026] Specifically, the system also includes a remote monitoring platform for displaying the location, liquid level, operation log, and system alarm information of the mobile water replenishment unit in real time.

[0027] A control method for a follow-up water supply system includes the following steps: Receive water replenishment request signals and real-time location information sent by the fire-fighting robot; The mobile water replenishment unit's travel path is planned based on the location information, and path following control is initiated. It detects obstacles in real time during its journey and performs obstacle avoidance maneuvers accordingly. Upon reaching the designated location, the automatic water replenishment process is initiated based on the water replenishment demand signal. Real-time monitoring of liquid level status, and sending an early warning signal when the liquid level is lower than a set threshold; After water replenishment is complete, it receives a stop command and enters standby mode, waiting for the next water replenishment task.

[0028] In this embodiment, the mobile water replenishment unit is the core execution carrier of the following water replenishment system based on the intelligent fire extinguishing robot for detecting and suppressing fires. Its core function is to load the fire extinguishing medium and have fully autonomous movement capabilities. It also integrates multiple functional components to adapt to the needs of following, obstacle avoidance, precise water replenishment, and multi-robot collaborative operation. Its overall structure and functional design are closely connected to the various units of the system, as detailed below: The mobile water replenishment unit uses a sealed liquid storage tank as the core for storing the extinguishing medium. Inside the tank, multiple sets of capacitive level sensors are arranged at preset intervals, and an ultrasonic level sensor is fixed at the top. The sensor array is attached to the inner wall of the tank via a stable mounting bracket, ensuring stable acquisition of capacitance data and top distance data at different heights even during movement, shaking, or foaming of the medium, providing a basic sensing source for the level monitoring unit. The outside of the tank integrates a booster pump, solenoid valve, flow sensor, pressure sensor, and pressure regulator. These components are connected sequentially via corrosion-resistant piping, extending to a water replenishment interface adapted to the fire extinguishing robot's water replenishment port. The interface uses a sealed quick-connect structure to ensure no medium leakage during replenishment and allows for rapid docking and disengagement with the fire extinguishing robot.

[0029] The autonomous movement of the mobile water replenishment unit is achieved by a dedicated drive system. This system comprises two drive motors and one steering motor, each connected to the unit's wheel system. The drive motors provide propulsion, while the steering motor controls the steering angle. Both motors feature rapid response capabilities, accurately receiving speed control commands, steering control commands, and turning angle adjustment commands from the path following control unit and obstacle avoidance detection unit. This enables the mobile water replenishment unit to move along the planned path, perform smooth obstacle avoidance maneuvers, and automatically return to the original path after obstacle avoidance. The drive system is also integrated with an inertial measurement unit (IMU), which collects the mobile water replenishment unit's acceleration and angular velocity information in real time, providing data support for the path following control unit's relative pose calculations.

[0030] The mobile water replenishment unit integrates an identity recognition module, which is deeply integrated with the unit control module. It can verify and pair with multiple firefighting robots through the communication unit, establish a dedicated collaborative operation link, realize orderly water replenishment management in scenarios where multiple robots are operating simultaneously, avoid water replenishment conflicts, and ensure that the water replenishment needs of each firefighting robot can be accurately responded to.

[0031] The mobile water replenishment unit is equipped with a dedicated sensor mounting structure, providing a stable mounting reference for various sensors required for path following and obstacle avoidance. The mounting structure secures a lidar, vision sensor, 2D lidar, depth camera, and ultrasonic sensor. The lidar and vision sensor capture 3D point cloud data of the traveling environment and visual feature tags for the firefighting robot. The 2D lidar acquires distance information to obstacles on the horizontal plane, the depth camera collects 3D scene information ahead, and the ultrasonic sensor detects low-lying obstacles at close range. The mounting positions of all sensors are calibrated to ensure that the perception range covers the entire traveling path. The collected data is transmitted to the path following control unit and obstacle avoidance detection unit via built-in wiring.

[0032] The mobile water replenishment unit is equipped with a control interface and establishes a fixed communication link with the central control unit. On the one hand, it receives control commands such as automatic start / stop, path adjustment, and water replenishment parameter adjustment issued by the central control unit. On the other hand, it provides real-time feedback on its own location information, movement status, liquid level data of the storage tank, water replenishment operation progress, and component operating status, ensuring the central control unit's overall coordination of the mobile water replenishment unit.

[0033] Overall, the mobile water replenishment unit achieves safe storage of fire extinguishing media through a storage tank, adapts to following and obstacle avoidance requirements through a drive system and sensor installation structure, achieves precise water replenishment and multi-robot collaboration through water replenishment execution components and an identification module, and integrates into the overall system control system through a control interface. Ultimately, it achieves fully autonomous operation of the entire process of "loading-following-obstacle avoidance-precise water replenishment-collaborative operation", providing stable fire extinguishing media replenishment support for the continuous operation of fire extinguishing robots.

[0034] In this embodiment, the path following control unit is the core control module that ensures the mobile water replenishment unit accurately tracks the fire extinguishing robot. Its core function is to receive the location information of the fire extinguishing robot and, relying on the path guidance mechanism and precision control algorithm, achieve high-precision, closed-loop following motion of the mobile water replenishment unit to the fire extinguishing robot. The unit consists of an environmental perception module, a data fusion module, and a motion control module. These modules work together and achieve following control through a systematic process, as detailed below: The environmental perception module is the fundamental data acquisition carrier for path-following control, integrating three core perception devices: LiDAR, vision sensors, and inertial measurement units. These three devices operate synchronously with consistent data acquisition timing. The LiDAR continuously scans the environment to acquire comprehensive 3D point cloud data, fully presenting the spatial structure and distance distribution of the surrounding environment. The vision sensor captures image information of the area ahead, its core function being to identify pre-set visual feature tags on the surface of the firefighting robot, enabling initial tracking of the robot through image localization of these feature tags. The inertial measurement unit focuses on acquiring the motion parameters of the mobile water replenishment unit itself, specifically acceleration and angular velocity data during movement, providing a motion reference for subsequent relative motion state calculations.

[0035] The data fusion module is responsible for processing and fusing multi-source data, a crucial step in achieving high-precision positioning. First, it performs spatiotemporal registration on the 3D point cloud data and image information collected by the environmental perception module, ensuring both types of data maintain a unified benchmark in both time and space dimensions, eliminating deviations caused by asynchronous data acquisition. Then, a filtering algorithm is used to process the registered dataset, removing noise data from the environment, including interference from lighting changes and point cloud errors caused by obstacle reflections. The purified effective data is used to accurately extract the real-time contour and spatial coordinates of the fire-fighting robot. Based on this, the data fusion module fuses the extracted fire-fighting robot position coordinates with its own acceleration and angular velocity data collected by the inertial measurement unit, and uses an extended Kalman filter algorithm for state estimation. Finally, it calculates the high-precision relative pose of the mobile water replenishment unit relative to the fire-fighting robot. This relative pose includes three key parameters: the straight-line distance between the mobile water replenishment unit and the fire-fighting robot, their azimuth angle in the horizontal plane, and the height difference in the vertical direction, providing a precise position reference for subsequent motion control.

[0036] The motion control module is responsible for converting relative pose data into specific execution commands to achieve dynamic following adjustments. The module has a pre-set following distance threshold, which is set according to the characteristics and safety requirements of the firefighting operation scenario, and simultaneously receives real-time speed data from the firefighting robot. Based on the relative pose output by the data fusion module, combined with the following distance threshold and the firefighting robot's speed, the motion control module dynamically calculates using a model predictive control algorithm to generate speed control and steering control commands for the mobile water replenishment unit. The speed control command specifies the magnitude of the mobile water replenishment unit's travel speed, ensuring it matches the firefighting robot's speed and maintains the preset following distance; the steering control command determines the steering angle, ensuring the mobile water replenishment unit always faces the firefighting robot and corrects directional deviations during travel.

[0037] The speed and steering control commands generated by the motion control module are sent to the drive system of the mobile water replenishment unit in real time. The drive system then drives the walking mechanism to move along the planned path according to the commands. Throughout the following process, the path following control unit continuously and cyclically executes the entire process of data acquisition, spatiotemporal registration, filtering, feature extraction, data fusion, state estimation, and command generation, forming a closed-loop feedback control mechanism. Each cycle dynamically adjusts the control commands based on the latest environmental data and motion status, promptly correcting following deviations caused by changes in the firefighting robot's motion and environmental interference. This ensures that the mobile water replenishment unit maintains a high-precision following state until it reaches the target water replenishment location and completes the following task.

[0038] The obstacle avoidance and detection unit is the core module that ensures the safe passage of the mobile water replenishment unit while following the firefighting robot. Its core function is to detect various obstacles in the path in real time. Through the coordinated process of environmental perception, obstacle recognition, path replanning, and motion control, the mobile water replenishment unit can achieve smooth turning and precise obstacle avoidance while maintaining a stable following state. The unit consists of a multimodal sensor array, an obstacle recognition and classification module, a dynamic path replanning module, and a motion execution module. These modules work together in an orderly manner and operate in a fixed logical cycle, as detailed below: A multimodal sensor array forms the perception foundation for obstacle avoidance and detection. It integrates three core devices: a 2D LiDAR, a depth camera, and an ultrasonic sensor. These three devices are installed in a pre-defined layout at the front and sides of the mobile water replenishment unit, ensuring the perception range covers the direction of travel and surrounding key areas. The 2D LiDAR focuses on acquiring distance information of obstacles in the horizontal plane, accurately providing feedback on the straight-line distance between obstacles in different directions and the mobile water replenishment unit by continuously scanning the areas in front and to the sides. The depth camera is responsible for collecting 3D spatial information of the scene in front, clearly presenting the three-dimensional structure, height, and spatial distribution of obstacles. The ultrasonic sensor focuses on detecting low-lying obstacles at close range, compensating for the perception blind spots of the LiDAR and depth camera in low-lying areas. All three devices start simultaneously and maintain a consistent data acquisition frequency, achieving comprehensive coverage and complementarity of multi-source perception data. It should be noted that the preset following distance threshold is 1.5-3m, and the first-level liquid level warning threshold is 30% of the water tank volume, and the second-level... 1. Warning threshold is 15%; "Fuzzy PID control proportional coefficient Kp=2.5, integral time Ti=0.8s, derivative time Td=0.3s"; 2. Supplement sensor performance parameters: "Lidar scanning frequency is 10Hz, detection distance is 0.1-50m"; "Vision sensor resolution is 1920×1080, frame rate is 30fps"; "Liquid level sensor measurement accuracy is ±1%FS"; 3. Supplement typical working condition test data: "In a dense smoke environment (visibility 5m), following accuracy error ≤±0.2m; water supply flow control accuracy ±5L / min, pressure control accuracy ±0.1MPa; obstacle avoidance response time ≤0.5s, obstacle avoidance success rate ≥98%".

[0039] The obstacle recognition and classification module is responsible for multi-source data processing and obstacle analysis. First, it preprocesses the multi-source sensing data collected by the sensor array, aligning timestamps and unifying spatial coordinates to eliminate differences in data acquisition timing and spatial reference deviations from different devices, ensuring data consistency. Then, it fuses the preprocessed multi-source data to generate an environmental grid map. This map uses grids as basic units to clearly identify static obstacle areas along the travel path. Simultaneously, through comparative analysis of continuous data frames, it tracks and identifies the movement trajectories of dynamic obstacles. Based on the environmental grid map, the module uses image segmentation and point cloud clustering algorithms for refined obstacle recognition, extracting the obstacle's contour features, actual size, and motion state. Then, according to preset judgment rules, all identified obstacles are divided into three categories: passable, avoidable, and dangerous / emergency. Passable obstacles are small obstacles that do not affect travel; avoidable obstacles affect the path but can be avoided by adjusting direction; and dangerous / emergency obstacles are high-risk obstacles that may pose a collision risk.

[0040] The dynamic path replanning module is the core component for obstacle avoidance decision-making, activating only when obstacles requiring avoidance or those posing a danger or emergency are identified. The module acquires the current pose of the mobile water replenishment unit in real time, including its current coordinates and direction of travel. Simultaneously, it receives the target location information and obstacle distribution data from the firefighting robot. Using these three as core input parameters, it performs real-time path calculation using the artificial potential field method. This method simulates a virtual force field, treating the mobile water replenishment unit as a point mass, attracting the target and repelling obstacles. By calculating the force equilibrium direction of the point mass within the force field, a collision-free local obstacle avoidance path is planned. After path planning, the module simultaneously generates corresponding turning angle and speed adjustment commands. The turning angle command specifies the direction the mobile water replenishment unit needs to adjust to ensure obstacle avoidance. The speed adjustment command adjusts the travel speed appropriately based on obstacle distance and path complexity, ensuring the smoothness of the obstacle avoidance maneuver.

[0041] The motion execution module is responsible for translating the control commands generated by the path replanning module into actual actions. After receiving turning angle and speed adjustment commands, the module precisely sends them to the steering and drive motors of the mobile water replenishment unit. The steering motor adjusts the wheel steering according to the turning angle command, and the drive motor changes its output power according to the speed adjustment command, controlling the mobile water replenishment unit to smoothly perform obstacle avoidance maneuvers while maintaining its following position on the firefighting robot. During obstacle avoidance, the module continuously receives real-time data from the sensor array and dynamically corrects the motion parameters to ensure the accuracy of the obstacle avoidance path. Once the mobile water replenishment unit successfully bypasses an obstacle, the motion execution module automatically controls the device to return to the originally planned following path, ensuring the continued execution of subsequent following tasks.

[0042] The entire obstacle avoidance and detection unit operates in a continuous cycle, from sensor data acquisition, preprocessing, fusion, identification, and classification to decision-making, planning, and execution. It repeats this process continuously until the mobile water replenishment unit reaches the target water replenishment location. Throughout the process, the mobile water replenishment unit avoids all potential collision risks while following the firefighting robot, ensuring safe movement.

[0043] The water replenishment control unit is the core functional module that enables the mobile water replenishment unit to accurately and on-demand replenish water to the firefighting robot. Its core function is to receive the water replenishment request signal from the firefighting robot, and through signal analysis, parameter calculation, dynamic regulation, and process control, achieve precise matching of water replenishment flow and pressure. This ensures that the water replenishment process is safe, stable, and meets the actual needs of firefighting operations. The unit consists of a water replenishment interface signal processing module, an adaptive control module, and an actuator. All components work together and complete the water replenishment operation according to a standardized process, as detailed below: The water replenishment interface is the physical connection carrier between the water replenishment control unit and the fire extinguishing robot. It adopts a sealed quick-connect structure, which has the characteristics of fast connection and strong sealing. It can quickly form a closed water replenishment channel after the mobile water replenishment unit docks with the fire extinguishing robot, effectively preventing the leakage of fire extinguishing medium during the water replenishment process. At the same time, the interface is compatible with the water replenishment port specifications of the fire extinguishing robot, ensuring smooth and unobstructed medium transmission and providing a reliable physical basis for subsequent accurate water replenishment.

[0044] The signal processing module is responsible for receiving and parsing the water replenishment demand signal. The module receives the water replenishment demand signal from the fire-fighting robot in real time through the communication unit, decodes and preprocesses the signal, and extracts key water replenishment parameters, including the target water replenishment flow rate, target water replenishment pressure, and estimated water replenishment duration. These parameters serve as the core basis for subsequent water replenishment control. The signal processing module verifies the validity of the parameters, eliminates abnormal data, and ensures the accuracy of the input parameters. The verified parameters are then transmitted to the adaptive control module.

[0045] The adaptive control module is the decision-making core of the water replenishment control unit, responsible for calculating and dynamically adjusting water replenishment parameters and managing the process. The module first acquires the current liquid level and internal pressure data of the mobile water replenishment unit's storage tank. It then combines this data with the target water replenishment flow rate and target water replenishment pressure transmitted from the signal processing module, and substitutes it into a preset fluid dynamics model to calculate the target opening degree of the solenoid valve, the target output pressure of the pressure regulating device, and the initial speed of the booster pump, providing initial control parameters for starting the water replenishment operation.

[0046] During the water replenishment operation, the adaptive control module continuously receives real-time data from the flow and pressure sensors—the actual flow rate and pressure values ​​flowing through the pipeline. It compares these actual parameters with the target parameters in real time and calculates the deviation. Based on this deviation, the module uses a fuzzy PID control algorithm for dynamic adjustment. By changing the opening of the solenoid valve to regulate the working state of the pressure regulating device and adjusting the speed of the booster pump, it achieves precise control of the water replenishment flow and pressure. This ensures that the actual flow and pressure remain stable within the target range, guaranteeing a high degree of match between the water replenishment effect and the operational requirements of the fire-fighting robot.

[0047] Meanwhile, the adaptive control module monitors two key status information in real time: the operational status reported by the fire-fighting robot and the liquid level status signal transmitted by the liquid level monitoring unit. If the fire-fighting robot's fire-fighting operation is suspended, or the liquid level monitoring unit triggers a low liquid level alarm, the module immediately activates the protection mechanism, entering a water replenishment pause or termination process to avoid equipment failure caused by ineffective water replenishment or media depletion. When the water replenishment time reaches the estimated water replenishment time, or when a stop water replenishment signal is received from the fire-fighting robot, the module issues control commands in a preset sequence to orderly shut down the solenoid valves and booster pumps, ensuring a smooth end to the water replenishment process and avoiding pipeline impact or media residue caused by sudden pressure changes.

[0048] The actuator, the execution vehicle of the water replenishment control unit, consists of a solenoid valve, a booster pump, a flow sensor, a pressure sensor, and a pressure regulating device. These components are connected sequentially via corrosion-resistant piping, forming a complete water replenishment transmission link. The solenoid valve receives opening commands from the adaptive control module, precisely controlling the valve opening degree and adjusting the medium flow cross-section. The booster pump starts and adjusts its operating state according to initial speed commands and dynamic adjustment commands, providing power for the transmission of the fire extinguishing medium. The flow and pressure sensors collect real-time flow and pressure data of the medium in the pipeline, continuously feeding back to the adaptive control module, forming the sensing basis for closed-loop control. The pressure regulating device, according to the target output pressure requirements, assists in stabilizing the pressure environment within the pipeline, ensuring the stability of the water replenishment process.

[0049] The entire water replenishment control unit obtains the required parameters through the signal processing module, completes the decision-making and regulation through the adaptive control module, and implements the action through the actuator. The three work together to form a complete water replenishment control system. The entire process is automated from docking and starting regulation to termination, which not only meets the real-time water replenishment needs of the fire extinguishing robot, but also ensures the safety and accuracy of the water replenishment operation.

[0050] The liquid level monitoring unit is the core sensing and control module that monitors the storage status of the fire extinguishing medium within the mobile water replenishment unit in real time, ensuring the continuous and stable operation of water replenishment. Its core functions are to accurately collect liquid level data, analyze the status level, trigger early warning signals, and link the system to execute corresponding strategies. The unit consists of a sensor array data processing module, a status assessment module, and an early warning linkage module. These modules operate collaboratively in a continuous loop, as detailed below: The sensor array serves as the fundamental carrier for liquid level data acquisition, employing a combination of multiple capacitive liquid level sensors and one ultrasonic liquid level sensor. Multiple capacitive liquid level sensors are vertically arranged at preset intervals on the inner wall of the storage tank, covering different height positions from the bottom to the top of the tank, and are used to synchronously collect capacitance data of the extinguishing medium at each height. The ultrasonic liquid level sensor is fixedly installed at the center of the top of the storage tank, emitting detection signals vertically downwards to collect distance data from the sensor probe to the top of the liquid surface. Both types of sensors start synchronously and maintain a consistent data acquisition frequency, ensuring comprehensive capture of the spatial distribution and overall height information of the liquid level within the tank, overcoming the measurement limitations of a single sensor under complex operating conditions.

[0051] The data processing module is responsible for data purification and accurate calculation. First, it receives capacitance data and top ranging data transmitted from the sensor array. It then cleans these two types of raw data, removing outliers caused by equipment errors or environmental interference. Next, it weights and fuses the cleaned, valid data, assigning different weights based on the measurement accuracy of the two types of sensors. Finally, it uses a filtering algorithm to effectively eliminate measurement noise caused by factors such as shaking during the movement of the mobile water supply unit and foam generated by the flow of the extinguishing medium. Finally, it accurately calculates the current liquid level and corresponding volume of the extinguishing medium through data conversion, providing accurate and reliable basic data for subsequent condition assessment.

[0052] The status assessment module is responsible for determining and classifying the liquid level status. The module has pre-set multiple dynamic thresholds. These thresholds are not fixed values ​​but are adaptively adjusted based on factors such as the current operational stage of the firefighting robot (e.g., high-intensity firefighting or routine inspection) and historical water replenishment records (e.g., single water replenishment volume and frequency). This ensures that the threshold settings are highly matched to the actual operational scenario. The status assessment module compares the precise liquid level output from the data processing module with the multi-level dynamic thresholds in real time, classifying the liquid level status into three categories: normal, alert, and emergency. A normal state indicates sufficient medium reserves to meet subsequent water replenishment needs; an alert state indicates that the medium reserves have dropped to the warning range, requiring preparation for replenishment; and an emergency state indicates that the medium reserves are nearing depletion and cannot continue to support water replenishment operations.

[0053] The early warning and linkage module is a crucial component for achieving status response and system coordination. When the status assessment module determines that the liquid level has entered a warning or emergency state, the early warning and linkage module immediately initiates the early warning signal generation process. The generated low liquid level early warning signal includes three core pieces of information: the current precise liquid level value, the estimated remaining operating time based on the current water replenishment rate and the remaining medium volume, and suggested actions for the current state, such as proceeding to the replenishment point or terminating water replenishment. After the early warning signal is generated, it is transmitted in real time to the central control unit and the firefighting robot via the communication unit, ensuring that relevant parties are promptly aware of any abnormal liquid level situations.

[0054] Simultaneously, the early warning linkage module establishes a linkage control mechanism with the central control unit, coordinating and executing corresponding strategies based on the level of the early warning signal. In a state of alert, the central control unit coordinates with the path-following control unit to plan the optimal route to the resupply point, while also coordinating with the water replenishment control unit to appropriately reduce the water replenishment flow rate and extend the remaining media's usability. In an emergency, the central control unit immediately issues an instruction to coordinate with the water replenishment control unit to terminate the water replenishment operation, and simultaneously coordinates with the path-following control unit to initiate an automatic return process, ensuring the mobile water replenishment unit's safe return for resupply.

[0055] The entire liquid level monitoring unit operates in a continuous cycle. From the sensor array collecting data to the data processing module refining and calculating, to the status assessment module determining the level, and the early warning linkage module generating signals and executing strategies, the operation is repeated continuously until the overall fire extinguishing and water replenishment task of the system is completely completed, ensuring accurate monitoring and dynamic control of the fire extinguishing medium liquid level throughout the process.

[0056] The communication unit is the core information interaction hub connecting the mobile water replenishment unit and the remote monitoring platform of the central control unit of the fire extinguishing robot. Its core function is to establish a stable signal transmission link, realize multi-directional real-time bidirectional data transmission, ensure the accurate transmission and synchronization of various key signals, and provide reliable communication support for the collaborative operation of various units in the system.

[0057] The communication unit adopts wireless communication protocols supporting multiple communication methods such as Wi-Fi, Bluetooth, and LoRa. It can adaptively select the appropriate communication method based on the environmental complexity and communication distance requirements of the firefighting operation scenario. Wi-Fi communication is suitable for short-range, high-bandwidth scenarios, ensuring rapid transmission of large amounts of data and meeting the high timeliness requirements of scenarios such as path-following control and precise water replenishment. Bluetooth communication is suitable for short-range, low-power device docking scenarios, enabling rapid identification and pairing signal interaction between the mobile water replenishment unit and the firefighting robot. LoRa communication is suitable for long-range, low-power scenarios, possessing strong anti-interference capabilities and maintaining stable transmission in complex fire environments such as smoke-covered areas, avoiding signal interruption. The overall transmission latency of the communication unit is low, and the signal transmission rate and stability are adapted to the system's operational requirements, ensuring real-time interaction of various critical commands and data without significant lag.

[0058] The communication unit undertakes the transmission of multiple types of signals throughout the entire process, covering all key aspects of the system operation. Firstly, it enables direct signal interaction between the mobile water replenishment unit and the fire-fighting robot. The fire-fighting robot transmits water replenishment demand signals to the mobile water replenishment unit, including target water replenishment flow rate, target water replenishment pressure, estimated water replenishment duration, and its own real-time position coordinates, movement speed, and operational status. The mobile water replenishment unit transmits liquid level status signals to the fire-fighting robot, including current extinguishing medium level, liquid volume, liquid level status level, low level warning details, and its own real-time position tracking status. Secondly, it connects the mobile water replenishment unit with the central control unit, transmitting liquid level warning signals, equipment operating status data, and communication link status information to the central control unit. Simultaneously, it receives automatic start / stop commands, path adjustment commands, and anomaly handling commands from the central control unit, ensuring the central control unit's global coordination of the mobile water replenishment unit. Thirdly, it works with the identification module to complete signal transmission, conveying equipment identification, pairing requests, and collaborative operation authorization signals between the mobile water replenishment unit and the fire-fighting robot, ensuring that the mobile water replenishment unit can be orderly paired with multiple fire-fighting robots to avoid operational conflicts. Its four-way remote monitoring platform transmits data including the real-time location information and liquid level status data of the mobile water replenishment unit, as well as complete operation logs and system alarm information, providing comprehensive visualization support for equipment operation for remote management personnel.

[0059] The communication unit is deeply integrated with various functional units of the system to form a closed-loop communication system, ensuring the targeted and effective transmission of signals. It directly connects with the path-following control unit to transmit the fire-fighting robot's position information in real time, providing core data input for the path-following control unit's environmental perception, relative pose calculation, and control command generation. It collaborates with the water replenishment control unit to accurately transmit the fire-fighting robot's water replenishment demand signals and operational status feedback signals, ensuring the water replenishment control unit can start and adjust water replenishment parameters in a timely manner. It links with the liquid level monitoring unit to quickly transmit low liquid level early warning signals, enabling the central control unit and the fire-fighting robot to respond and execute corresponding strategies promptly. It works with the identity recognition module to ensure identity verification and pairing communication in multi-robot collaborative operations. It maintains a continuous and stable communication link with the remote monitoring platform to ensure real-time uploading of equipment operation data and timely early warning of abnormal situations.

[0060] The communication unit incorporates a built-in communication status monitoring and interruption recovery mechanism to adapt to the anomaly handling requirements of the central control unit. The unit monitors the connection status of the communication link in real time, including key indicators such as signal strength, transmission delay, and data packet loss rate. When a communication interruption is detected, an automatic reconnection process is automatically initiated. It first attempts to restore the connection using the original communication method. If multiple reconnections fail, it automatically switches to a backup communication method based on preset communication method priorities to ensure uninterrupted transmission of core signals such as water replenishment requests, liquid level warnings, and location information. During the reconnection process, the communication unit temporarily buffers critical signal data, retransmitting it immediately after the link is restored to avoid data loss affecting system operations. Simultaneously, the communication unit feeds back the communication anomaly status to the central control unit in real time so that the system can adjust its operational strategy promptly to ensure the continuity of the water replenishment task.

[0061] The communication unit operates in an automated, cyclical manner. Upon system startup, the communication unit automatically initializes, establishes an initial communication link with the remote monitoring platform of the fire-fighting robot's central control unit, and completes device identity verification. During operation, signals to be transmitted by each unit are collected at a fixed frequency, encoded, and then accurately transmitted via the selected communication method. Simultaneously, feedback signals from each end are received in real time, decoded, and quickly distributed to the corresponding functional units. The communication status is dynamically monitored throughout the process, adaptively adjusting transmission parameters based on environmental changes and link conditions to ensure the accuracy, real-time nature, and stability of signal transmission until the system completes all water replenishment tasks. The communication link is then closed in an orderly manner according to instructions.

[0062] The central control unit is the core hub of the follow-up water replenishment system based on the intelligent fire extinguishing robot for detecting and suppressing fires. Its core function is to coordinate the orderly operation of various functional units of the mobile water replenishment unit, realize the automatic start and stop control, real-time status management, multi-unit collaborative linkage and emergency response of the entire system process, and ensure the efficient, safe and stable execution of the follow-up water replenishment task.

[0063] The central control unit houses the main control module, which serves as the core of the entire unit's computation and instruction set. This module integrates a high-performance processor and a dedicated control chip, enabling rapid data processing, multi-task parallel scheduling, and precise instruction generation. Upon system startup, the main control module first completes the overall system initialization process, performing power-on self-tests on the path following control unit, obstacle avoidance detection unit, water replenishment control unit, liquid level monitoring unit, communication unit, and the drive system and identification module of the mobile water replenishment unit. It verifies the hardware status, communication link connectivity, and initial parameter configuration of each unit. After confirming no abnormalities, it issues a system startup command, and each unit simultaneously enters a standby state. During task execution, the main control module schedules each unit to work collaboratively according to a preset sequence, ensuring the consistency of data acquisition, instruction transmission, and action execution. After the task is completed, it receives a stop command and systematically issues shutdown commands to each unit, completing system status reset and data archiving.

[0064] The central control unit includes a collaborative scheduling module, specifically responsible for coordinating the linkage of various functional units and constructing a closed-loop collaborative mechanism. During the following phase, the collaborative scheduling module receives the fire-fighting robot's position information and speed data in real time via the communication unit, and synchronously sends this information to the path-following control unit, instructing it to initiate multi-sensor fusion positioning and path planning. Simultaneously, it continuously receives environmental grid maps, obstacle classification results, and path replanning suggestions uploaded by the obstacle avoidance detection unit. When obstacle avoidance is required, the coordinated path-following control unit suspends its original path instructions and prioritizes executing the turning angle and speed adjustment instructions generated by the obstacle avoidance detection unit. After obstacle avoidance is completed, the path-following control unit is instructed to return to the original planned path, ensuring seamless integration of following and obstacle avoidance actions. During the water replenishment phase, the collaborative scheduling module receives the fire-fighting robot's water replenishment request signal and synchronously forwards it to the water replenishment control unit, instructing it to initiate water replenishment parameter calculation and actuator operation. Simultaneously, it acquires the current liquid level data and status assessment results from the liquid level monitoring unit in real time, matches the liquid level information with the water replenishment request, and sends dynamic flow and pressure adjustment suggestions to the water replenishment control unit, ensuring that the water replenishment process is compatible with the media storage capacity.

[0065] The central control unit is equipped with a status management module that collects and integrates the operational status data of the entire system in real time. This module continuously receives status feedback from each unit via a communication link, including relative pose data and following accuracy deviation values ​​from the path-following control unit; sensor operating status and obstacle recognition results from the obstacle avoidance detection unit; solenoid valve opening, booster pump speed, and actual flow and pressure data from the water replenishment control unit; liquid level height, medium volume, and liquid level status level from the liquid level monitoring unit; link signal strength, transmission delay, and data packet loss rate from the communication unit; and drive motor speed, steering angle, and remaining battery power from the mobile water replenishment unit. The status management module categorizes, organizes, and updates this data in real time, constructing a system operational status matrix to provide data support for subsequent scheduling decisions and anomaly detection. Simultaneously, it uploads key status data to a remote monitoring platform via the communication unit, enabling system status visualization.

[0066] The central control unit has a built-in anomaly handling module that integrates multi-scenario anomaly response strategies to ensure system reliability under complex operating conditions. When a path loss signal is received from the path following control unit, the anomaly handling module immediately instructs the path following control unit to restart the environmental perception and data fusion process, re-extract the fire-fighting robot's position coordinates, and quickly replan the following path using data from the inertial measurement unit. If multiple replanning attempts fail, the module coordinates with the obstacle avoidance detection unit to expand the perception range and assist in locating the fire-fighting robot. When the communication unit reports a communication interruption, the anomaly handling module instructs the communication unit to initiate an automatic reconnection process and switch to a backup communication mode. During reconnection, each unit is instructed to maintain its current operating state and temporarily cache critical data, retransmitting the data once the link is restored. If reconnection fails for an extended period, a safety mechanism is activated, instructing the mobile water replenishment unit to reduce its travel speed and prioritize executing the nearest return-to-base or standby command. When the liquid level monitoring unit sends a low liquid level warning signal, the anomaly handling module executes corresponding strategies based on the warning level: In alert mode, it instructs the water replenishment control unit to reduce the water replenishment flow rate, while simultaneously instructing the path following control unit to plan the optimal replenishment path based on the current location and replenishment point distribution; in emergency mode, it immediately instructs the water replenishment control unit to terminate the water replenishment operation, shut down the solenoid valve and booster pump, and simultaneously instructs the path following control unit to initiate an automatic return process to ensure the safe return of the mobile water replenishment unit. When the water replenishment control unit reports that the flow rate or pressure deviation exceeds the threshold, the anomaly handling module instructs the adaptive control module to strengthen the fuzzy PID adjustment, while simultaneously verifying the liquid storage tank pressure and sensor status to eliminate potential hardware malfunctions. In addition, a priority scoring model is established based on the fire-fighting robot's remaining power (weight 60%), remaining water volume (weight 30%), and task urgency (weight 10%), with the highest-scoring robot receiving priority for water replenishment; 2. Supplemental communication degradation strategy: "When the signal interference intensity at the fire scene is ≥-85dBm, it automatically switches to LoRa communication mode, reducing the data transmission rate to 1kbps to ensure the transmission of core control commands"; 3. Emergency handling for water replenishment docking failure: "If docking fails 3 times (single docking time ≤5s), the following angle is automatically adjusted (±10°) and retried. If there are a total of 5 failures, an alarm signal is sent to the command center."

[0067] The central control unit is equipped with a remote interaction module, enabling bidirectional data transmission and command interaction with the remote monitoring platform. The remote interaction module receives control commands from the remote monitoring platform in real time, including manual start / stop of water replenishment, adjustment of follow distance thresholds, modification of liquid level warning thresholds, and designation of replenishment point locations. After verification by the main control module, these commands are converted into executable operation instructions for each unit. Simultaneously, the module uploads system operating status data, operation logs, and abnormal alarm information to the remote monitoring platform at a preset frequency. The operation logs include key information such as water replenishment time, water replenishment volume, follow path, obstacle avoidance attempts, and abnormal handling records, facilitating traceability and analysis by remote management personnel. When a serious abnormality is triggered in the system, the remote interaction module prioritizes uploading alarm information, including the abnormality type, occurrence time, and current equipment location, supporting rapid intervention and handling by remote management personnel.

[0068] The central control unit includes a pairing management module that works closely with the identification module of the mobile water replenishment unit to achieve multi-robot collaborative operation management. The pairing management module receives its own device identifier uploaded by the identification module of the mobile water replenishment unit, and simultaneously receives pairing requests and device identifiers from each firefighting robot via the communication unit. It verifies and matches the identifiers of both parties to establish a dedicated collaborative link. Based on the firefighting robot's operational priority, the urgency of water replenishment needs, and the current liquid level and location of the mobile water replenishment unit, it rationally allocates water replenishment resources, generates a pairing operation list, and instructs the mobile water replenishment unit to execute water replenishment tasks in the order of the list. When multiple firefighting robots issue water replenishment requests simultaneously, the pairing management module coordinates conflicts to avoid duplicate or missed water replenishment, ensuring orderly collaboration in multi-robot operation scenarios.

[0069] The central control unit operates in an automated closed-loop management mode. After system startup, it enters standby mode after initialization self-test; upon receiving the water replenishment request and location information of the fire extinguishing robot, the main control module coordinates with the scheduling module to initiate the following and obstacle avoidance process; upon reaching the water replenishment location, it instructs the water replenishment control unit to perform precise water replenishment; throughout the process, the operating status is monitored by the status management module, and the anomaly handling module responds to various emergencies in real time; when the task is completed or an emergency is triggered, the shutdown, return, or replenishment process is executed in an orderly manner; at the same time, the remote interaction module maintains real-time linkage with the monitoring platform until the entire system task is completed, the status is reset and the data is archived, all without manual intervention, realizing intelligent control of the entire following and water replenishment process.

[0070] In the description of this invention, it should be understood that the terms "coaxial," "bottom," "one end," "top," "middle," "other end," "upper," "side," "top," "inner," "front," "center," "both ends," etc., indicate the orientation or positional relationship based on the drawings, and are only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0071] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "setting," "connection," "fixing," "screw connection," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal connection of two components or the interaction between two components. Unless otherwise explicitly limited, those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0072] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that variations may be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A follow-up water replenishment system based on an intelligent fire-fighting robot for detecting and suppressing fires, characterized in that, include: Mobile water supply unit, used to load fire extinguishing medium and has autonomous mobility; The path following control unit is used to receive the position information of the fire extinguishing robot and realize the dynamic following of the fire extinguishing robot based on the multi-sensor fusion positioning algorithm, and control the mobile water replenishment unit to follow the movement of the fire extinguishing robot. The obstacle avoidance detection unit is used to detect obstacles in the travel path in real time and control the mobile water replenishment unit to turn and avoid obstacles; The water replenishment control unit is used to adjust the water replenishment flow rate and pressure according to the water replenishment demand signal of the fire extinguishing robot through a fuzzy PID control algorithm, and control the mobile water replenishment unit to automatically replenish water to the fire extinguishing robot. The liquid level monitoring unit is used to monitor the liquid level of the fire extinguishing medium in the mobile water replenishment unit in real time, and to realize adaptive determination of water replenishment timing by combining the graded dynamic liquid level threshold, and generate liquid level status signal. The communication unit is used to realize the signal interaction between the mobile water replenishment unit and the fire extinguishing robot, including the transmission of water replenishment demand signals, liquid level status signals and location information. The central control unit establishes bidirectional signal connections with the path following control unit, the obstacle avoidance detection unit, the water replenishment control unit, and the communication unit, respectively. It receives real-time data collected by each unit and outputs control commands to coordinate the work of each unit, thereby realizing automatic start-stop control and system status management of the mobile water replenishment unit.

2. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, as described in claim 1, is characterized in that: The path-following control unit employs a multi-sensor fusion positioning algorithm that integrates data from lidar, visual sensors, and inertial navigation to achieve high-precision following of the firefighting robot by the mobile water replenishment unit. The path-following control unit includes an environmental perception module, a data fusion module, and a motion control module, and achieves high-precision following through the following steps: Step S1: Obtain three-dimensional point cloud data of the traveling environment by scanning with lidar, capture image information in front by visual sensor to identify visual feature tags of the fire extinguishing robot, and obtain the acceleration and angular velocity information of the mobile water replenishment unit itself by inertial measurement unit. Step S2: Perform spatiotemporal registration of the three-dimensional point cloud data and the image information, remove environmental noise data through a filtering algorithm, and extract the real-time contour and position coordinates of the fire-fighting robot; Step S3: Combine the position coordinates of the fire-fighting robot with the data of the inertial measurement unit, and use the extended Kalman filter algorithm to perform state estimation to obtain the high-precision relative pose of the mobile water replenishment unit relative to the fire-fighting robot, including distance, azimuth angle and height difference; Step S4: Based on the relative pose, and according to the preset following distance threshold and the movement speed of the fire extinguishing robot, the speed control command and steering control command of the mobile water replenishment unit are dynamically generated through the model predictive control algorithm. Step S5: Send the speed control command and steering control command to the drive system of the mobile water replenishment unit, so that it moves along the planned path, and continuously repeat steps S1 to S4 during the movement to form a closed-loop feedback control until the following task is completed.

3. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, as described in claim 1, is characterized in that: The obstacle avoidance detection unit includes an ultrasonic sensor, an infrared sensor, and a visual recognition module, which can build an environmental map in real time and plan a collision-free path. The obstacle avoidance detection unit includes a multimodal sensor array, an obstacle recognition and classification module, a dynamic path replanning module, and a motion execution module, and achieves real-time obstacle avoidance and turning control through the following steps: Step P1: Synchronously collect multi-source perception data of the traveling environment through the multimodal sensor array, wherein the two-dimensional lidar acquires the distance information of obstacles on the horizontal plane, the depth camera acquires the three-dimensional scene information in front, and the ultrasonic sensor detects low obstacles at close range. Step P2: The multi-source sensing data is time-stamped and aligned with the spatial coordinate system, fused to generate an environmental grid map, and the static obstacle area and the dynamic obstacle trajectory are marked in the map. Step P3: Based on the environmental grid map, the outline, size and motion state of the obstacles are identified by image segmentation and point cloud clustering algorithms, and the obstacles are classified into traversable, avoidable and dangerous / emergency types according to preset rules. Step P4: When an obstacle that needs to be avoided or is in danger or emergency is identified, a collision-free local obstacle avoidance path is calculated in real time using the artificial potential field method based on the current pose of the mobile water replenishment unit, the target position of the fire extinguishing robot and the distribution of obstacles, and corresponding turning angle and speed adjustment commands are generated. Step P5: Send the turning angle and speed adjustment commands to the steering motor and drive motor of the mobile water replenishment unit, control it to perform smooth obstacle avoidance while maintaining the following state, and automatically return to the original planned path after obstacle avoidance. The whole process is continuously and cyclically executed until the mobile water replenishment unit reaches the target water replenishment position.

4. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, as described in claim 1, is characterized in that: The water replenishment control unit includes a solenoid valve, a flow sensor, and a pressure regulating device, which can adjust the water replenishment flow and pressure according to the real-time fire extinguishing medium requirements of the fire extinguishing robot. The water replenishment control unit includes a water replenishment interface, a signal processing module, an adaptive control module, and an actuator, and achieves precise on-demand water replenishment through the following steps: Step C1: Receive water replenishment demand signals from the fire extinguishing robot in real time through the communication unit. After parsing, the signals include the target water replenishment flow rate, the target water replenishment pressure, and the estimated water replenishment duration. Step C2: Obtain the current liquid level data and internal pressure data of the mobile water replenishment unit's storage tank, and combine them with the target water replenishment flow rate and target water replenishment pressure. Calculate the target opening degree of the solenoid valve, the target output pressure of the pressure regulating device, and the initial speed of the booster pump using a preset fluid dynamics model. Step C3: Control the solenoid valve to open to the target opening degree, and simultaneously start the booster pump to make the extinguishing medium flow through the flow sensor and pressure sensor; Step C4: Collect the actual flow rate and pressure value of the pipeline in real time, compare it with the target flow rate and target pressure, and use the fuzzy PID control algorithm to dynamically adjust the opening of the solenoid valve, the state of the pressure regulating device and the speed of the booster pump to stabilize the actual flow rate and pressure within the target range. Step C5: During the water replenishment process, continuously monitor the operational status fed back by the fire-fighting robot. If its fire-fighting operation is suspended or the liquid level monitoring unit triggers a low liquid level alarm, immediately enter the protective water replenishment suspension or termination process. When the estimated water replenishment time is reached or a stop water replenishment signal is received from the fire-fighting robot, orderly shut down the solenoid valve and booster pump to complete this water replenishment operation.

5. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires, as described in claim 1, is characterized in that: The liquid level monitoring unit uses a capacitive liquid level sensor or an ultrasonic liquid level sensor to monitor the liquid level in real time and trigger a low liquid level warning signal. The liquid level monitoring unit includes a sensor array, a data processing module, a status assessment module, and an early warning linkage module, and achieves intelligent liquid level monitoring and early warning through the following steps: Step L1: Simultaneously collect capacitance data and top distance data of the fire extinguishing medium at different heights in the tank by using multiple sets of capacitive liquid level sensors and one set of ultrasonic liquid level sensors arranged in the storage tank. Step L2: Perform data cleaning and weighted fusion on the capacitance data and ranging data, and use a filtering algorithm to eliminate measurement noise caused by movement, shaking or medium foam, and calculate the accurate height value of the current liquid level and the medium volume. Step L3: Compare the precise liquid level height with a preset multi-level dynamic threshold. The multi-level dynamic threshold is adaptively adjusted according to the current operation stage of the fire extinguishing robot and the water replenishment history, and the liquid level status is determined to be normal, warning or emergency based on this. Step L4: When the liquid level enters a warning or emergency state, a low liquid level warning signal is generated, which includes the current liquid level value, the estimated remaining working time and the suggested operation. The warning signal is then sent to the central control unit and the fire extinguishing robot in real time through the communication unit. Step L5: The central control unit coordinates the path following control unit and the water replenishment control unit to execute corresponding strategies based on the received warning signal level. These strategies include reducing the water replenishment flow and planning the optimal path to the replenishment point in the alert state, or immediately terminating water replenishment and initiating an automatic return process in the emergency state. The entire monitoring process continues to cycle until the system task is completed.

6. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires according to claim 1, characterized in that: The communication unit adopts a wireless communication protocol and supports real-time bidirectional data transmission, including but not limited to Wi-Fi, Bluetooth or LoRa communication methods.

7. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires according to claim 1, characterized in that: The central control unit is also equipped with an anomaly handling mechanism, including path loss replanning, automatic reconnection after communication interruption, liquid level anomaly alarm, and automatic return-to-home function.

8. The following water replenishment system based on an intelligent fire extinguishing robot for detecting and suppressing fires according to claim 1, characterized in that: The mobile water replenishment unit is also equipped with an identification module for pairing and collaborative operation management with multiple firefighting robots.

9. A water replenishment system based on a fire detection and suppression intelligent fire extinguishing robot according to claim 1, characterized in that: The system also includes a remote monitoring platform for displaying the location, liquid level, operation log, and system alarm information of the mobile water replenishment unit in real time.

10. A control method based on the follow-up water replenishment system according to any one of claims 1-9, characterized in that, Includes the following steps: Receive water replenishment request signals and real-time location information sent by the fire-fighting robot; The mobile water replenishment unit's travel path is planned based on the location information, and path following control is initiated. It detects obstacles in real time during its journey and performs obstacle avoidance maneuvers accordingly. Upon reaching the designated location, the automatic water replenishment process is initiated based on the water replenishment demand signal. Real-time monitoring of liquid level status, and sending an early warning signal when the liquid level is lower than a set threshold; After water replenishment is complete, it receives a stop command and enters standby mode, waiting for the next water replenishment task.

Images