Service robot with external emergency power supply capability and control method thereof

Abstract

This invention relates to the fields of robotics applications and power electronics technology, and discloses a service robot with external emergency power supply capability and its control method. The robot includes a load characteristic identification module, used to analyze load parameters by injecting micro-perturbation pulses, and extract the equivalent capacitance only when the load is determined to be in capacitive cold start mode; a path energy consumption inversion module, used to analyze the terrain elevation of the return path, calculate the peak driving power required to pass through the steepest area using a dynamic model, and deduce the minimum reserved energy to ensure safe homing; a thermodynamic verification module, used to establish the correlation constraint between battery thermal effect and return dynamic performance, and calculate the allowable power supply; and a flexible power supply control module, used to respond to power supply triggering, perform soft start based on preceding parameters, and cut off power supply when the battery level is low. This invention can suppress startup surges and ensure that the remaining energy and power after the robot is powered on meet the homing requirements in complex terrain.

Description

Technical Field

[0001] This invention relates to the fields of robot applications and power electronics technology, specifically to a service robot with external emergency power supply capability and its control method. Background Technology

[0002] With the development of mobile robot technology, the application scenarios of service robots have expanded from indoor environments to complex outdoor parks and wilderness environments. In addition to routine inspection and transportation tasks, the ability to provide emergency power has become an important functional extension for specialized service robots (such as rescue robots and mobile charging robots). These robots are typically equipped with high-energy-density battery packs, capable of providing temporary power support to external devices that are running low on power.

[0003] In practical applications, existing power supply control strategies primarily focus on power conversion efficiency and basic overcurrent protection, but they fall short in managing system safety and survivability under complex operating conditions. Regarding energy management, current technologies typically use a fixed remaining power threshold (e.g., retaining 10% power) as the basis for cutting off power. However, this static strategy does not consider the robot's current geographical location or the terrain features of its return path. When the robot is in a low-altitude area and needs to climb a slope to return, or when the return path has high-resistance terrain, the fixed percentage of power retained is often insufficient to overcome terrain resistance, causing the robot to run out of energy and be unable to return autonomously.

[0004] Regarding electrical connections, the powered devices that service robots encounter are often diverse and have unknown electrical characteristics. If the input terminal of the powered device has a large-capacity filter capacitor that is depleted, directly closing the main circuit to supply power will instantly generate a huge surge current. This surge can easily trigger the robot's short-circuit protection mechanism, leading to power supply failure, and in severe cases, it can even damage the power devices inside the DC-DC converter or the input circuit of the powered device.

[0005] Furthermore, high-power external power supply inevitably leads to an increase in battery temperature. Battery management systems typically automatically limit battery discharge power based on temperature rise to ensure safety. Existing control schemes often overlook the relationship between this battery thermal effect and robot dynamic performance. If the robot continues to operate at full power, causing the battery to overheat, even with sufficient remaining charge, the limited battery discharge power may not be enough to meet the instantaneous drive power requirements of the robot during its return climb, causing the robot to be unable to pass through obstacle sections due to power limitations on the return journey. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a service robot with external emergency power supply capability and its control method. This solves the problems in existing technologies where service robots lack dynamic energy baseline calculation based on return terrain, do not establish a soft-start mechanism for large capacitor loads, and ignore the constraint of thermal derating caused by power supply on return climbing power, resulting in problems such as homing failure, hardware damage, and power interruption.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a service robot with external emergency power supply capability. The service robot includes a load feature identification module, a path energy consumption inversion module, a thermodynamic verification module, and a flexible power supply control module.

[0008] The load characteristic identification module is used to analyze the electrical parameters on the load side and determine the load type. In the initial stage of power supply circuit closure, this module injects a low-voltage, high-frequency micro-disturbance signal into the load circuit as a detection pulse through an active detection unit and collects time-domain data.

[0009] The impedance analysis unit performs frequency domain transformation on the time-domain data to calculate the equivalent resistance, equivalent capacitance, and equivalent inductance on the load side. Based on these parameters, the module executes load type determination logic: if the calculated equivalent resistance is higher than the short-circuit threshold, the equivalent capacitance value is not within the measurement noise range, and the equivalent inductance value is within the measurement noise range, the load type is determined to be a capacitive load cold start mode. In this case, the equivalent capacitance is extracted as a key parameter and a flexible power supply operation is triggered. If the equivalent resistance is higher than the short-circuit threshold and both capacitance and inductance are noise values, the load type is determined to be a standard resistive load mode and a regular power supply operation is triggered. If the equivalent resistance is too low or there is significant inductance, the load type is determined to be a fault mode and the power output is locked.

[0010] The path energy consumption inversion module is used to plan the optimal return path based on the service robot's spatial location and environmental data, and to calculate the minimum reserved energy and driving peak power to constrain the power supply process.

[0011] This module analyzes environmental map data through a topology analysis unit, extracts terrain elevation information for the optimal return path, and identifies areas with maximum slope and / or high resistance by analyzing the elevation change rate. The module uses a dynamic model to calculate the peak driving power required for the robot to traverse these difficult areas fully loaded. This dynamic model decomposes the driving force into the slope component that overcomes gravity and the ground friction resistance. The topology analysis unit calculates the downward component of gravity along the slope and the normal force component perpendicular to the slope, and combines this with the ground friction coefficient to derive the total running resistance. The product of this total running resistance and the preset climbing speed is then used as the peak driving power.

[0012] Meanwhile, the module calculates the total return distance through path integration and combines it with the fence calculation unit to calculate the minimum reserve energy. The calculation logic for the minimum reserve energy is as follows: the product of the total return distance, average driving energy consumption, and safety redundancy coefficient is used as the dynamic energy part, and the product of the basic static power consumption and return time is used as the static energy part. The sum of the two is the minimum reserve energy to ensure a safe return home.

[0013] The thermodynamic verification module is used to establish the correlation constraint between the battery thermal effect and the return power performance of the service robot. It calculates the allowable power supply to constrain the power supply process by taking into account the real-time battery temperature and the aforementioned peak drive power.

[0014] This module uses the battery internal resistance thermal model through the temperature prediction unit to calculate the predicted temperature value after a preset evaluation window time for continuous power supply at the requested rated power, and then queries the temperature and power discharge table of the battery management system based on the predicted temperature value to obtain the maximum discharge power.

[0015] The derating decision unit calculates the allowable power supply based on the relationship between the maximum discharge power and the peak drive power: when the maximum discharge power is greater than the sum of the power safety margin and the peak drive power, the requested rated power is taken as the allowable power supply; when the maximum discharge power is insufficient, the reverse iterative approximation method is used to monotonically decrease the requested rated power and substitute it into the model for recalculation until a derating power value is found that allows the new maximum discharge power at that value to meet the return power requirements. This derating power value is the allowable power supply.

[0016] The flexible power supply control module is used to respond to the triggering of flexible power supply operation, and performs soft start to supply power to the outside based on the extracted equivalent capacitance and the calculated allowable power supply.

[0017] The strategy generation unit first converts the allowable power supply into the maximum allowable current, then uses the equivalent capacitance to calculate the start-up time constant according to the time constant formula, and generates a target voltage curve that conforms to the exponential growth law based on the start-up time constant and the rated voltage of the equipment.

[0018] The monitoring unit drives the bidirectional DC-DC converter to change its output voltage according to the target voltage curve. During stable power supply, the module executes monitoring and power cut-off logic based on the minimum reserved energy: when the remaining battery energy is equal to or lower than the minimum reserved energy, or the battery temperature rise rate exceeds the safety threshold, or the real-time load current is lower than the no-load threshold, the power cut-off logic is executed.

[0019] A second aspect of the present invention provides a control method for a service robot with external emergency power supply capability, the method comprising the following steps: Analyze the electrical parameters on the load side and determine the load type; calculate the equivalent resistance, equivalent capacitance, and equivalent inductance by injecting probe pulses into the load circuit and collecting time-domain data; when the load type is determined to be a capacitive load cold start mode, extract the equivalent capacitance and trigger flexible power supply operation.

[0020] Plan the optimal return path and calculate the minimum reserved energy and peak driving power to constrain the power supply process; by identifying the maximum slope area or high resistance area in the path, calculate the peak driving power required for the robot to pass through the area based on the dynamic model, and calculate the minimum reserved energy to ensure safe return based on the return distance and energy consumption parameters.

[0021] Establish a constraint relationship between battery thermal effect and service robot return trip dynamic performance; comprehensively consider real-time battery temperature and peak drive power, and calculate the allowable power supply to constrain the power supply process by predicting the battery temperature after power supply and the corresponding maximum discharge capacity.

[0022] In response to the triggering of flexible power supply operation, it performs power supply control; it performs soft start to supply power to the outside based on the equivalent capacitance and allowable power supply power, and performs monitoring and power cut-off logic based on the minimum reserved energy during stable power supply.

[0023] This invention provides a service robot with external emergency power supply capability and its control method. It has the following beneficial effects: 1. This invention achieves quantitative calculation of the service robot's return journey survivability by constructing an energy inversion mechanism based on geographic topology and dynamics models. By analyzing the terrain elevation information in the return path, the system can accurately identify areas with the maximum slope or high resistance, and use the dynamics model to calculate the peak driving power required to pass through the area. Combined with the total return distance and the minimum reserved energy calculated by the dynamic and static energy consumption models, a dynamically adjustable energy threshold is set for external power supply, effectively preventing the robot from exhausting the energy required for the return journey due to excessive power supply, and ensuring the robot's autonomous homing capability in complex terrain.

[0024] 2. This invention proposes a flexible start-up strategy based on load impedance characteristic identification, which solves the surge impact problem caused by direct power-on of large capacitor loads. By injecting a low-voltage, high-frequency micro-perturbation signal before power supply, the system can accurately identify the equivalent capacitance of the load under the safe state of main circuit shutdown. Based on the start-up time constant calculated from the capacitance value and allowable power, a voltage curve conforming to the exponential growth law is generated, realizing precise control of the output voltage. This soft-start method keeps the charging current within a safe range, avoiding overcurrent protection or hardware damage caused by surge current, and improving the safety and success rate of the power supply process.

[0025] 3. This invention establishes a coupling constraint mechanism between battery thermal effect and robot dynamic performance, ensuring the return trip dynamic performance after the power supply task. The system does not simply determine the power supply power based on the current battery state, but predicts the battery temperature at the end of the power supply and checks the maximum discharge power at that temperature. Through reverse iterative calculation, the system can dynamically adjust the external power supply power to ensure that although the battery temperature rises after the power supply ends, its discharge capacity is still sufficient to cover the peak drive power and safety margin required for the return trip. This effectively avoids the situation where the power is limited due to battery overheating, resulting in insufficient power or inability to move during the return trip. Attached Figure Description

[0026] Figure 1 This is a service robot architecture diagram with external emergency power supply capability according to an embodiment of the present invention; Figure 2 This is an architecture diagram of a load feature identification module according to an embodiment of the present invention; Figure 3 This is a diagram of the architecture of a thermodynamic verification module according to an embodiment of the present invention; Figure 4 This is a flowchart illustrating a control method for a service robot with external emergency power supply capability according to an embodiment of the present invention. Figure 5 Voltage and current simulation waveforms of a flexible soft-start process according to an embodiment of the present invention; Figure 6 A battery state prediction curve for the thermal and dynamic coupling verification process according to an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions in 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.

[0028] Please see the appendix Figure 1This invention provides a service robot with emergency power supply capability. The service robot's hardware includes a robot main control system, a bidirectional DC-DC converter, a pulse generator circuit, an analog-to-digital converter circuit, a SLAM navigation system, a battery management system, a current sensor, and environmental sensors. The robot main control system, as the core scheduling unit, establishes communication connections with the bidirectional DC-DC converter, the SLAM navigation system, and the battery management system via an industrial fieldbus to issue commands and transmit status data. The bidirectional DC-DC converter, as a power interface unit, has a pulse generator circuit and an analog-to-digital converter circuit at its output to collaboratively achieve physical detection and high-frequency sampling of the electrical characteristics of the powered equipment.

[0029] The service robot may include a load feature identification module 100, a path energy consumption inversion module 200, a thermodynamic verification module 300, and a flexible power supply control module 400.

[0030] The load characteristic identification module 100 is used to identify the electrical parameters on the load side in the initial stage of power supply circuit closure; the path energy consumption inversion module 200 is used to inversely deduce the minimum energy required for survival based on the robot's spatial position and environmental data; the thermodynamic verification module 300 is used to establish the correlation constraint between battery thermal effect and robot return dynamic performance; the flexible power supply control module 400 is used to generate voltage output strategy and execute power supply operation based on upstream parameters.

[0031] The load characteristic identification module 100 is equipped with an active detection unit and an impedance analysis unit. After receiving a connection confirmation signal from the robot's main control system, the load characteristic identification module 100 controls the bidirectional DC-DC converter to maintain the main power off state and drives the pulse generation circuit to inject a detection pulse into the load circuit. This detection pulse is a low-voltage, high-frequency micro-perturbation signal.

[0032] The load characteristic identification module 100 uses an analog-to-digital converter circuit to acquire the response current waveform and terminal voltage of the load circuit, generating time-domain data. The impedance analysis unit performs a Fourier transform algorithm on the time-domain data to calculate the electrical parameters (equivalent resistance, equivalent capacitance, and equivalent inductance) of the load side. Based on the calculation results, the load characteristic identification module 100 determines the load type. If the load type is determined to be a capacitive load in cold start mode, the equivalent capacitance is extracted for subsequent control and a flexible power supply operation is triggered.

[0033] The path energy consumption inversion module 200 is equipped with a topology analysis unit and a fence calculation unit. The module acquires coordinate data and environmental map data through a SLAM navigation system, and plans the optimal return path from the current coordinate point to the nearest charging station. By parsing the environmental map data, the module extracts terrain elevation information, identifies the areas with the steepest slope or high resistance in the optimal return path, calculates the peak drive power required for the robot to pass through these areas fully loaded, and calculates the total return distance.

[0034] The path energy consumption inversion module 200 calculates the minimum retained energy based on the retained energy formula. The retained energy formula is as follows: ; In the formula: Minimum Reserved Energy refers to the minimum battery energy required to ensure the robot can safely return to the charging station. The total return distance refers to the path length planned from the current coordinate point to the nearest charging station; Average driving energy consumption refers to the average energy required to move one meter, which is calculated based on the robot's past driving data. The safety redundancy factor refers to a dimensionless constant preset to cope with uncertainties such as skidding and detours; Basic static power consumption refers to the power consumption of basic circuits and computing power when the robot is in a non-dormant state but not moving. Average return speed refers to the estimated average speed of movement of the robot when performing a homing task.

[0035] The thermodynamic calibration module 300 is equipped with a temperature prediction unit and a derating decision unit. The thermodynamic calibration module 300 collects the current ambient temperature from environmental sensors and the real-time battery temperature from the battery management system, and receives externally input requested rated power. The thermodynamic calibration module 300 uses a battery internal resistance thermal model to calculate the predicted temperature value after continuous power supply at the requested rated power for a certain period. Based on the predicted temperature value, the thermodynamic calibration module 300 queries the temperature-power discharge table of the battery management system to obtain the maximum discharge power.

[0036] The thermodynamic verification module 300 receives the maximum discharge power, the peak drive power calculated by the path energy consumption inversion module 200, and the requested rated power, and calculates the allowable power supply according to the allowable power formula. The allowable power formula is as follows: ; In the formula: Permissible power supply refers to the maximum power that the robot is ultimately allowed to output after thermodynamic verification. The requested rated power refers to the target power supply requested by the power receiving equipment or the user side. Maximum discharge power refers to the power at the predicted temperature. Below, the maximum power limit that the battery management system allows the battery to output; The predicted temperature value refers to the battery temperature at the moment of power supply termination calculated based on the thermal model. Peak power refers to the instantaneous driving power required for the robot to pass through the section with the greatest gradient or greatest resistance on the return path; For power safety margin, it refers to the power buffer value reserved to ensure the redundancy of the power system; The derating power value refers to the calculated reduced external power supply setting value when full power supply causes the battery to overheat and affect the return trip power.

[0037] The flexible power supply control module 400 is equipped with a strategy generation unit and an execution monitoring unit. In response to the triggering of flexible power supply operation, the flexible power supply control module 400 receives the equivalent capacitance output by the load characteristic identification module 100, the allowable power supply output by the thermodynamic verification module 300, and the rated voltage of the equipment. The flexible power supply control module 400 first converts the allowable power supply into the maximum allowable current, and then calculates the start-up time constant according to the time constant formula. The time constant formula is as follows: ; In the formula: The startup time constant refers to the time parameter that controls the rate of voltage rise during the voltage soft-start process. Rated voltage refers to the target voltage value required for the normal operation of the powered equipment. Equivalent capacitance refers to the equivalent capacitance at the input terminal of the power receiving device measured during the feature identification stage. The maximum permissible current refers to the current limit value calculated by dividing the allowable power supply by the rated voltage of the equipment.

[0038] The flexible power supply control module 400 generates a target voltage curve based on the startup time constant and the equipment's rated voltage using a voltage curve formula. The voltage curve formula is as follows: ; In the formula: The target voltage curve refers to the voltage over time. Theoretical output voltage value of a varying bidirectional DC-DC converter; is the base of the natural logarithm.

[0039] The flexible power supply control module 400 drives the bidirectional DC-DC converter to adjust the output voltage according to the target voltage curve. During the stable power supply phase, the flexible power supply control module 400 continuously monitors the battery's remaining energy, battery temperature rise rate, and real-time load current from the current sensor in the battery management system.

[0040] When the remaining battery energy is equal to or lower than the minimum retainable energy calculated by the path energy consumption inversion module 200 If the battery temperature rise rate exceeds the safety threshold, or the real-time load current is lower than the no-load threshold, the flexible power supply control module 400 cuts off the external power supply circuit and sends a homing trigger signal to the robot's main control system.

[0041] See appendix Figure 2 The load characteristic identification module 100 is used to identify the electrical characteristics of the powered equipment using low-energy pulse detection technology before establishing a high-power power supply path. The load characteristic identification module 100 includes an active detection unit and an impedance analysis unit. The active detection unit performs signal injection and acquisition, while the impedance analysis unit performs data processing and logical judgment.

[0042] The signal input terminal of the active detection unit is communicatively connected to the robot's main control system. Once the robot's main control system confirms that the physical interface (such as a charging gun or docking probe) is reliably connected to the powered device, it generates a connection confirmation signal. In response to this connection confirmation signal, the active detection unit first sends a latching command to the bidirectional DC-DC converter, forcing the power switching devices of the bidirectional DC-DC converter to remain off, thereby blocking the energy output of the main power circuit and preventing inrush current from being generated when power is directly applied under unknown load characteristics.

[0043] With the main circuit off, the active detection unit controls the pulse generation circuit to operate. The pulse generation circuit is configured to inject detection pulses into the load circuit consisting of the robot and the powered device.

[0044] The probe pulse is a preset low-voltage, high-frequency micro-perturbation signal. Its voltage amplitude is set within a safe voltage range (e.g., between 1V and 5V), and its frequency is set in the mid-to-high frequency range (e.g., between 1kHz and 10kHz). This safe voltage range is set below the conduction threshold or breakdown voltage of conventional power semiconductor devices (such as diodes and MOSFET body diodes), ensuring that the probe signal will not falsely trigger the power circuitry inside the powered device, thus achieving non-destructive detection.

[0045] Simultaneously, the active detection unit triggers the analog-to-digital converter (ADC) circuit to synchronously sample the load circuit. The ADC circuit acquires the terminal voltage and response current waveforms of the load circuit under the detection pulse excitation at a high sampling rate in the microsecond range (e.g., above 1 Msps) in real time, generating time-domain data. This time-domain data completely records the transient response characteristics of the load circuit to the excitation signal and is transmitted to the impedance analysis unit.

[0046] The impedance analysis unit receives the aforementioned time-domain data and uses a digital signal processor to execute a Fourier transform algorithm (Fast Fourier Transform, FFT or Discrete Fourier Transform, DFT).

[0047] Specifically, the impedance analysis unit samples the time-domain voltage signal. With time-domain current signal Transformed into frequency domain voltage phasors With frequency domain current phasor Calculate the complex impedance at the current frequency using Ohm's law in the frequency domain. The impedance analysis unit will analyze the complex impedance. The resistance is decomposed into real and imaginary parts, where the real part represents the equivalent resistance on the load side. The imaginary part represents the reactance component on the load side.

[0048] If the imaginary part is negative, it indicates that the load exhibits capacitive characteristics, and the equivalent capacitance can be calculated. If the imaginary part is positive, then the equivalent inductance can be analyzed. .

[0049] The specific computational process of the Fourier transform algorithm in discrete signal processing, including the selection of the window function and the handling of spectral leakage, is a well-known technique in the field of digital signal processing and will not be elaborated here.

[0050] The impedance analysis unit is based on the calculated load-side electrical parameters (equivalent resistance). Equivalent capacitance and equivalent inductance The system uses preset logical criteria to determine the load type. These criteria include three main modes: The first is the physical short-circuit mode or fault mode. If the calculated equivalent resistance... Less than or equal to the short-circuit threshold, and the equivalent capacitance With equivalent inductance If the value is within the measurement noise range or the system resolution limit (considered a minimum), the impedance analysis unit determines that there is a physical short circuit in the load circuit. The short circuit threshold is a safety lower limit set based on the sum of the inherent impedance of the service robot's external power supply cable and the standard contact resistance of the physical interface (e.g., it can be set to 0.1Ω). In this case, the impedance analysis unit immediately generates a system alarm signal and locks the bidirectional DC-DC converter, prohibiting power output.

[0051] The second type is the standard resistive load mode. If the calculated equivalent resistance... Above the short-circuit threshold, and the equivalent capacitance With equivalent inductance If the value is within the measurement noise range or the system resolution limit, it indicates that the load exhibits purely resistive characteristics (such as resistance heating equipment or lighting equipment without large capacitors). In this case, the impedance analysis unit determines it to be a standard resistive load and notifies the system to enter the normal power supply process.

[0052] The third type is the capacitive load cold start mode. If the calculated equivalent resistance... It exceeds the short-circuit threshold, but is accompanied by a significant equivalent capacitance. (e.g., exceeding 1000μF), and equivalent inductance If the value is within the measurement noise range or system resolution limit, it indicates that a large-capacity filter capacitor is connected in parallel at the input of the powered equipment and is in a depleted state. At this time, the impedance analysis unit determines it to be a capacitive load cold start mode. In this mode, the impedance analysis unit triggers a flexible power supply operation, extracting the equivalent capacitance... This parameter is locked as a key control parameter and passed to the flexible power supply control module 400 for the generation of subsequent soft-start strategies. This step ensures that the subsequent voltage ramp-up rate can accurately match the charging characteristics of the load capacitor.

[0053] Furthermore, for cases where the calculated equivalent resistance is less than or equal to the short-circuit threshold (regardless of whether it is accompanied by significant equivalent capacitance or equivalent inductance), and for cases where the calculated equivalent inductance value is not within the measurement noise range or system resolution limit (i.e. exhibiting significant inductive characteristics), the impedance analysis unit also classifies them as the first physical short-circuit mode or fault mode and performs protection actions.

[0054] On the one hand, when the equivalent series resistance or real impedance of the load is lower than the line safety threshold, regardless of whether it has capacitive or inductive characteristics, it means that the input terminal inside the power receiving equipment may have broken down and short-circuited, or that there is a serious insulation failure in the load circuit. On the other hand, the flexible power supply control strategy in this embodiment is designed based on the characteristics of resistor-capacitor charging. If the load is significantly inductive (e.g., a high-power motor or transformer coil without a freewheeling circuit), the back electromotive force generated at the moment the detection pulse is cut off or the power supply is started may exceed the withstand voltage limit of the bidirectional DC-DC converter power devices. To ensure the hardware safety of the service robot itself, power supply is strictly prohibited under the above conditions.

[0055] The path energy consumption inversion module 200 serves as the computational and decision-making unit for energy safety management of service robots, primarily based on geospatial information to dynamically quantify the robot's survival energy baseline. This module mainly consists of a topology analysis unit and a fence calculation unit.

[0056] The topology analysis unit maintains real-time data interaction with the service robot's SLAM navigation system. While the service robot is performing external power supply tasks, the topology analysis unit periodically retrieves current coordinate data and environmental map data from the SLAM navigation system. This environmental map data includes not only conventional two-dimensional grid occupancy information, but also terrain elevation information that reflects the undulations of the environment (such as a 2.5D grid map with height attributes or three-dimensional point cloud data).

[0057] The topology analysis unit, based on the current coordinates, environmental map data, and the coordinates of the nearest charging station pre-stored in the system database, uses a path planning algorithm (such as the A algorithm or Dijkstra's algorithm) to plan a theoretically optimal return path. The specific algorithm implementation for path planning is a well-known technology in the field of mobile robot navigation and will not be elaborated upon here.

[0058] The topology analysis unit further extracts terrain features from the planned optimal return path. By analyzing the elevation change rate at discrete points along the path, the topology analysis unit identifies areas of maximum slope or high resistance (e.g., rough terrain based on map semantic annotations). Based on the robot's dynamics model, the topology analysis unit calculates the peak driving power required for the robot to traverse these most challenging sections at full load. .

[0059] Specifically, this dynamic model decomposes the robot's driving force into the slope component that overcomes gravity, ground friction resistance, and air resistance (negligible at low speeds). The topology analysis unit calculates the component of gravity along the slope (proportional to the sine of the slope angle) and the normal force component perpendicular to the slope (proportional to the cosine of the slope angle) based on the identified slope angle, and then calculates the total running resistance by combining this with the ground friction coefficient. Peak driving power. This is the product of the total running resistance and the preset climbing speed. This calculation process ensures that the obtained power value can physically cover the robot's power requirements when encountering maximum terrain resistance during the return journey. Simultaneously, the topology analysis unit accurately calculates the total return distance by performing path integration on the optimal return path. .

[0060] The fence calculation unit performs quantitative calculations of the energy fence based on the geographical and dynamic parameters output by the topology analysis unit. This unit calls upon historical operating parameters stored in the system database, including average driving energy consumption. Basic static power consumption and average return speed .

[0061] Among them, average driving energy consumption This is a key parameter reflecting the robot's current mechanical efficiency and load status. This parameter is updated in real-time using a sliding time window algorithm, meaning the system only calculates the total energy consumption and total mileage within the most recent travel distance (e.g., the last 1 to 5 kilometers) and then calculates their ratio. This statistical method effectively eliminates interference from earlier data and accurately reflects energy consumption fluctuations caused by mechanical wear or load changes.

[0062] Basic static power consumption This refers to the total power consumption of the robot's main control system, sensor group, and communication module in standby mode. This value is usually determined by the system's factory calibration or obtained through static current monitoring.

[0063] Average return speed It is the desired speed set according to the robot's navigation strategy.

[0064] The fence calculation unit calculates the minimum retained energy based on the energy retention formula. This formula combines the dynamic energy required for movement and the static energy required for return travel, and introduces a redundancy factor to address environmental uncertainties. The energy retention formula is as follows: ; In the formula: Minimum Reserved Energy refers to the minimum battery energy required to ensure the robot can safely return to the charging station. The total return distance refers to the path length planned from the current coordinate point to the nearest charging station; Average driving energy consumption refers to the average energy required to move one meter, which is calculated based on the robot's past driving data. The safety redundancy factor is a dimensionless constant preset to cope with uncertainties such as slippage and detours. This safety redundancy factor is set according to the complexity of the application scenario (for example, it is set to 1.1 in a flat indoor warehouse environment and 1.3 in a complex outdoor park environment). Basic static power consumption refers to the power consumption of basic circuits and computing power when the robot is in a non-dormant state but not moving. Average return speed refers to the estimated average speed of movement of the robot when performing a homing task.

[0065] Based on the above calculations, the fence calculation unit derives a dynamically changing energy threshold, i.e., the minimum retention energy. This value is not fixed, but is updated in real time as the distance between the robot and the charging station changes and the terrain of the return path changes. This ensures that no matter where the robot is, it always has enough and necessary energy to support its return to the charging facility, thus realizing dynamic adjustment of energy management strategy based on geographical location and working conditions.

[0066] See appendix Figure 3 The thermodynamic calibration module 300 establishes a constraint relationship between the battery's thermophysical properties and the robot's kinematic requirements. This is to prevent the battery pack temperature from becoming too high due to external power supply, which could trigger the battery management system to forcibly limit the discharge power, ultimately preventing the robot from completing its return mission. This module mainly includes a temperature prediction unit and a derating adjudication unit.

[0067] The temperature prediction unit reads real-time ambient temperature data from environmental sensors and real-time battery temperature data reported by the battery management system via an internal bus. Simultaneously, this unit receives the rated power request transmitted by the powered device via a communication protocol handshake or preset by the user. Based on this data, the temperature prediction unit performs temperature rise prediction calculations using a battery internal resistance thermal model.

[0068] Specifically, the battery internal resistance thermal model is built based on the principle of thermal balance. The temperature prediction unit first calculates the estimated discharge current based on the requested rated power and the current total battery voltage. Then, according to the law of conservation of energy, the heat accumulation inside the battery is equal to the difference between the rate of heat generation and the rate of heat dissipation.

[0069] The heat generation rate is calculated by multiplying the square of the estimated discharge current by the battery's DC internal resistance (following Joule's law); the heat dissipation rate is calculated by dividing the temperature difference between the battery's real-time temperature and the current ambient temperature by the equivalent thermal resistance of the heat dissipation system (following Newton's law of cooling). The temperature prediction unit performs discrete integration on the above heat accumulation process within a preset evaluation window (e.g., 10 to 30 minutes) to calculate the battery temperature rise curve over time. The final temperature calculated is the predicted temperature value.

[0070] The temperature prediction unit further calls upon a pre-set temperature-power discharge table in the battery management system. This table is a multi-dimensional lookup table whose data comes from the specification sheets or offline charge-discharge test data of individual battery cells. It stores the maximum continuous discharge power allowed by the battery's physical and chemical characteristics under different temperature ranges and different states of charge (SOC). Using the calculated predicted temperature value and the current SOC as index keys, the temperature prediction unit retrieves the physical boundary value that the battery is allowed to output at the end of the power supply, i.e., the maximum discharge power, from this table.

[0071] The derating decision unit receives the maximum discharge power from the temperature prediction unit, the peak drive power from the path energy consumption inversion module 200 (specifically the topology analysis unit), and the requested rated power. This unit introduces a power safety margin as an evaluation buffer, which is a power value reserved based on the power consumption and aging degradation of the robot-assisted system (e.g., set to 300W to 500W, covering the power consumption of the computing unit, sensor power consumption, and motor efficiency degradation margin).

[0072] The derating decision unit compares the maximum discharge power with the sum of the drive peak power and the power safety margin to determine whether the battery state after power supply is sufficient to support the most difficult sections of the return trip.

[0073] The derating decision-making unit calculates the final allowable power supply based on the allowable power formula. The allowable power formula is as follows: ; In the formula: Permissible power supply refers to the maximum power that the robot is ultimately allowed to output after thermodynamic verification. The requested rated power refers to the target power supply requested by the power receiving equipment or the user side. Maximum discharge power refers to the power at the predicted temperature. Below, the maximum power limit that the battery management system allows the battery to output; The predicted temperature value refers to the battery temperature at the moment of power supply termination calculated based on the thermal model. Peak power refers to the instantaneous driving power required for the robot to pass through the section with the greatest gradient or greatest resistance on the return path; For power safety margin, it refers to the power buffer value reserved to ensure the redundancy of the power system; The derating power value refers to the calculated reduced external power supply setting value when full power supply causes the battery to overheat and affect the return trip power.

[0074] For the derating power value in the above formula The specific calculations for derating are performed by the derating decision unit using a reverse iterative approximation method. The specific iterative process is as follows: the derating decision unit monotonically reduces the requested rated power in a set step size (e.g., 50W), and then substitutes the reduced power value back into the battery internal resistance thermal model of the temperature prediction unit to recalculate the corresponding new predicted temperature value and the new maximum discharge power. This process is repeated until a critical power value is found, such that the new maximum discharge power at this power value is greater than the sum of the drive peak power and the power safety margin. This critical power value is then determined as the derating power value. .

[0075] Through this logic, the system can ensure that even when the battery heats up due to external power supply, the remaining discharge capacity of the battery is always higher than the power threshold required for the return climb, thus ensuring at the physical level that the robot's return power performance is not impaired by the power supply task.

[0076] The flexible power supply control module 400 is used to convert the physical constraint parameters calculated by the preceding modules into specific power electronic control commands, so as to realize dynamic adjustment of the external output voltage and safety protection during the power supply process. The flexible power supply control module 400 mainly consists of a strategy generation unit and an execution monitoring unit.

[0077] The strategy generation unit is responsible for calculating the voltage output parameters, and its input terminal receives the equivalent capacitance from the load feature identification module 100. Allowable power supply from thermodynamic verification module 300 And the device's rated voltage from external communication protocol handshakes or user-defined settings. .

[0078] The strategy generation unit first performs current boundary calculation, dividing the allowable power supply by the equipment's rated voltage to calculate the maximum allowable current. This maximum permissible current characterizes the upper limit of current that the system can provide to the load under the constraints of the current battery thermal state and return power demand.

[0079] Based on the charging characteristics of capacitors in physics, the current flowing through a capacitor is directly proportional to the rate of change of voltage, that is... If a step voltage is directly output to a capacitive load, the rate of voltage change will approach infinity instantaneously, resulting in a huge inrush current. Therefore, the strategy generation unit calculates the startup time constant based on the time constant formula to smooth the voltage rise slope.

[0080] Specifically, this embodiment uses an exponential voltage rise curve. For the objective function... Its voltage change rate (slope) The rate of change occurs at the instant of initiation (i.e., When it reaches its maximum value, it is In order to limit the maximum inrush current during startup. Within this range, the following conditions must be met: Based on this, the formula for the time constant is derived as follows: ; In the formula: The startup time constant refers to the time parameter that controls the rate of voltage rise during the voltage soft-start process. Rated voltage refers to the target voltage value required for the normal operation of the powered equipment. Equivalent capacitance refers to the equivalent capacitance at the input terminal of the power receiving device measured during the feature identification stage. The maximum permissible current refers to the current limit value calculated by dividing the allowable power supply by the rated voltage of the equipment.

[0081] The strategy generation unit uses the calculated startup time constant to construct a soft-start strategy that conforms to an exponential growth law, namely the voltage curve formula. This curve simulates the natural charging process of a resistor-capacitor circuit, ensuring that the charging current flowing through the load capacitor during startup is always clamped within the maximum allowable current range. The voltage curve formula is as follows: ; In the formula: The target voltage curve refers to the voltage over time. Theoretical output voltage value of a varying bidirectional DC-DC converter; is the base of the natural logarithm.

[0082] The execution monitoring unit receives the target voltage curve generated above. It is then used as a reference signal input to the underlying control loop (e.g., PID controller) of the bidirectional DC-DC converter.

[0083] The execution monitoring unit adjusts the duty cycle of the power switch transistors of the bidirectional DC-DC converter to drive the actual output voltage to strictly follow the target voltage curve, thereby completing a smooth soft-start process.

[0084] When the actual output voltage reaches a preset percentage (e.g., 95% or 98%) of the device's rated voltage, the soft start is considered complete, and the system enters constant voltage power supply mode. The specific implementation of the PID control algorithm in power supply voltage tracking is a well-known technology in the field of power electronics control and will not be elaborated upon here.

[0085] After the soft start is complete and the power supply is stable, the monitoring unit switches to real-time safety monitoring. This unit continuously reads the remaining battery energy and battery temperature rise rate reported by the battery management system, and simultaneously reads the real-time load current collected by the current sensor. The monitoring unit executes three types of protection disconnection logic: Firstly, the energy cutoff logic. The monitoring unit compares the remaining battery energy with the minimum reserve energy from the path energy consumption inversion module 200 (specifically, the fence calculation unit) in real time. Once the remaining battery energy drops to or below the minimum reserve energy, indicating that the current power level has reached the critical point for the robot's survival during return, the monitoring unit will immediately trigger a shutdown.

[0086] Secondly, the thermal runaway prevention and shutdown logic. The execution monitoring unit monitors the battery temperature rise rate and compares it with a preset safety threshold. This safety threshold is an upper limit of the temperature rise rate set based on the chemical characteristics of the battery cell and the critical conditions for thermal runaway (e.g., set to 2℃ / min or 3℃ / min). If the battery temperature rise rate exceeds this safety threshold, it indicates a sharp increase in internal heat generation and a risk of thermal runaway, and the execution monitoring unit immediately triggers shutdown.

[0087] Third, the no-load disconnection logic. The execution monitoring unit monitors the real-time load current and compares it with a preset no-load threshold. This no-load threshold is a lower limit set based on the standby current characteristics of the powered device (e.g., set to 5% of the rated current). If the real-time load current remains below this no-load threshold for a preset duration (e.g., 3 seconds), it indicates that the powered device is fully charged or the connection has been disconnected, and the execution monitoring unit triggers a disconnection to avoid unnecessary standby power loss.

[0088] When any of the above-mentioned cutoff conditions are triggered, the execution monitoring unit immediately controls the bidirectional DC-DC converter to stop working and cuts off the external power supply circuit through a physical relay. At the same time, the execution monitoring unit sends a homing trigger signal to the robot's main control system, driving the robot to use its reserved energy to execute an automatic homing procedure, ensuring that the robot can safely return to the charging station.

[0089] See appendix Figure 4 The present invention also proposes a control method for a service robot with external emergency power supply capability, the method comprising the following steps: Step S1: Active identification and mode determination of load characteristics. The load characteristic identification module 100 receives a connection confirmation signal from the robot's main control system and controls the bidirectional DC-DC converter to maintain the main power off state. The load characteristic identification module 100 drives the pulse generation circuit to inject probe pulses into the load circuit and uses the analog-to-digital converter circuit to collect time-domain data. The load characteristic identification module 100 uses the Fourier transform algorithm to process the time-domain data and calculates the equivalent resistance, equivalent capacitance, and equivalent inductance on the load side. The load characteristic identification module 100 determines the load type based on the calculation results. If the equivalent resistance is less than or equal to the short-circuit threshold, or the equivalent inductance shows obvious inductive characteristics, it is determined to be a fault mode and the output is prohibited. If the equivalent resistance is higher than the short-circuit threshold and the equivalent capacitance and equivalent inductance are extremely small, it is determined to be a standard resistive load mode and enters the normal startup process. If the equivalent resistance is higher than the short-circuit threshold and is accompanied by a significant equivalent capacitance, it is determined to be a capacitive load cold start mode, and the equivalent capacitance is retained as a key parameter.

[0090] Step S2: Constructing an energy fence based on the return topology. The path energy consumption inversion module 200 retrieves coordinate data and environmental map data from the SLAM navigation system in real time to plan the optimal return path to the nearest charging station.

[0091] The path energy consumption inversion module 200 analyzes map data to extract terrain elevation information for the optimal return path, identifies areas with maximum slope or high resistance in the path, and calculates the peak drive power required for the robot to pass through the area fully loaded, as well as the total return distance. Combining the robot's average driving energy consumption, basic static power consumption, and average return speed, the path energy consumption inversion module 200 calculates the minimum energy required to ensure a safe return.

[0092] Step S3: Power boundary verification for thermal and dynamic coupling. The thermal dynamic verification module 300 collects the current ambient temperature and the real-time battery temperature, and calculates the predicted temperature value at the end of power supply using the battery internal resistance thermal model, in conjunction with the requested rated power. Based on this predicted temperature value, the thermal dynamic verification module 300 queries the temperature-power discharge table of the battery management system to obtain the maximum discharge power. The thermal dynamic verification module 300 compares the maximum discharge power with the sum of the drive peak power and the power safety margin. If the maximum discharge power is sufficient, the rated power output is permitted. If insufficient, a reduced derating power value is calculated through reverse iteration as the allowable power supply to prevent insufficient return power due to battery overheating.

[0093] Step S4: Flexible Start-up Execution and Survival Monitoring. The flexible power supply control module 400 receives the equivalent capacitance extracted in step S1, the allowable power supply determined in step S3, and the rated voltage of the equipment. The flexible power supply control module 400 calculates the start-up time constant and generates a target voltage curve that conforms to an exponential growth law. The flexible power supply control module 400 drives the bidirectional DC-DC converter to output voltage according to the target voltage curve, completing the soft start.

[0094] During periods of stable power supply, the flexible power supply control module 400 monitors the remaining battery energy, battery temperature rise rate, and real-time load current in real time, and executes energy cut-off logic, thermal runaway prevention cut-off logic, and no-load cut-off logic: once the remaining battery energy reaches the minimum reserve energy, the battery temperature rise rate is abnormal, or the equipment is disconnected, the power supply circuit is immediately physically cut off and a homing signal is triggered.

[0095] See appendix Figure 5 and attached Figure 6 To verify the effectiveness of the technical solution of the present invention in complex industrial environments, this embodiment constructs a specific scenario for automated operation and maintenance of a large data center.

[0096] In this scenario, a service robot needs to provide emergency power to a distributed edge computing gateway (powered device) that cannot start due to a power outage. The service robot is pre-equipped with a 48V / 20Ah lithium iron phosphate battery pack. The current battery temperature is 30℃, the ambient temperature is 25℃, the robot's full-load weight is 50kg, and the basic static power consumption is... The average driving energy consumption is 50W under flat road conditions. The rated voltage of the receiving equipment is 15Wh / km. 24V, rated power It has a power rating of 150W and its input terminal is connected in parallel with a large-capacity filter capacitor of up to 50mF (i.e., 50000μF).

[0097] In the initial phase of the power supply task, after the service robot establishes a physical connection with the powered equipment, the load feature identification module 100 immediately performs active detection. The system first injects a 5kHz, 1V probe pulse into the load circuit, acquires the response signal through an analog-to-digital converter circuit, and analyzes it via FFT transformation to calculate the equivalent resistance on the load side. Approximately 3.84Ω (corresponding to a steady-state impedance of 24V / 150W), equivalent inductance It is approximately 0H, while the equivalent capacitance is... The measured value is 50.2mF. Given that the equivalent resistance is greater than the system's set short-circuit threshold of 0.1Ω and there are obvious capacitive characteristics, the system determines that the current load is in "capacitive load cold start mode" and uses 50.2mF as the key input parameter for subsequent control.

[0098] Meanwhile, the path energy consumption inversion module 200 evaluates the return path based on data from the SLAM navigation system. The system plans a total path distance of 500 meters to the nearest charging station, including a 100-meter uphill section with a 10° gradient at the end of the path, which constitutes the maximum power demand condition for the return trip. Based on the dynamic model calculation, the peak drive power required for the robot to pass through this ramp at a speed of 0.5 m / s at full load is approximately 400W. Combining the energy retention formula, the system calculates the minimum energy retention to be 150Wh. Since the current remaining battery energy is 300Wh, which meets the minimum energy requirement for external power supply, the system is allowed to proceed to the next stage of verification.

[0099] Subsequently, the thermal dynamics verification module 300 performs the core power boundary calculation. Based on the user-requested 150W power supply and the expected 20-minute power supply duration, the system uses a battery internal resistance thermal model to extrapolate the calculations. It finds that if full power is supplied, the battery temperature will rise from 30℃ to 55℃ at the end of the power supply. Consulting the battery management system's temperature-power-discharge table reveals that at 55℃, to prevent the battery temperature from becoming too high, the maximum discharge power of the battery is... The power supply will be limited to 350W. However, the sum of the peak drive power required for the robot's return trip (400W) and the reserved power safety margin (50W) is 450W. At this point, there is a risk that the discharge power will be limited due to battery temperature rise, thus failing to meet the return trip drive power requirements. To address this, the system performs a reverse iterative calculation and ultimately decides to dredge the allowable power supply to 100W. At this power, the predicted battery temperature at the end of the power supply is only 45°C, and the corresponding battery management system allows a discharge power of 500W, which is sufficient to cover the 450W return trip power requirement, thereby ensuring that the battery state after the power supply ends meets the return trip dynamics constraints.

[0100] Finally, the flexible power supply control module 400 generates an execution strategy based on the determined physical constraints. The system first converts the allowable power supply of 100W into the maximum allowable current at 24V. The current is approximately 4.16A. Combining this with the measured equivalent capacitance of 50.2mF, the start-up time constant is calculated using the time constant formula. Approximately 0.288 seconds. The system generates the target voltage curve based on this. It also drives the bidirectional DC-DC converter to perform.

[0101] Appendix Figure 5 As can be seen from the simulation waveform, the voltage in this embodiment follows an exponential curve, with a rise time of about 1.5 seconds. During this period, the current is always controlled below 4.2A, and the overcurrent protection is not triggered.

[0102] At the same time, attached Figure 6The battery state prediction curve shows that, under the derating power supply strategy, the battery temperature trajectory (curve A) ends at 45°C. The maximum discharge power allowed by the battery management system at this temperature (curve B) is always higher than the return power demand threshold (curve C).

[0103] The above results demonstrate that the present invention can ensure the safe start-up of a robot with large capacity loads while ensuring the robot has reliable return capability.

Claims

1. A service robot with external emergency power supply capability, characterized in that, include: The load feature identification module (100) is used to analyze the electrical parameters on the load side and determine the load type, and extracts the equivalent capacitance and triggers the power supply operation only when the load type is determined to be a capacitive load cold start mode. The path energy consumption inversion module (200) is used to plan the optimal return path based on the spatial location and environmental data of the service robot, and to reverse-determine the minimum reserved energy and driving peak power required for the service robot to safely return home. Thermodynamic verification module (300) is used to establish the correlation constraint between battery thermal effect and service robot return power performance, and calculate the allowable power supply for constraining the power supply process by combining real-time battery temperature and the drive peak power. The flexible power supply control module (400) is used to respond to the triggering of the flexible power supply operation, perform soft start to supply power to the outside based on the equivalent capacitance and the allowable power supply power, and perform monitoring and power cut-off logic based on the minimum reserved energy during stable power supply.

2. A service robot with external emergency power supply capability according to claim 1, characterized in that, The load feature identification module (100) includes an active detection unit; The active detection unit is used to inject detection pulses into the load circuit and collect time-domain data during the initial stage of power supply circuit closure. The probe pulse is a low-voltage, high-frequency micro-perturbation signal, and the voltage amplitude of the probe pulse is set to a safe voltage range lower than the turn-on voltage of the power semiconductor device.

3. A service robot with external emergency power supply capability according to claim 2, characterized in that, The load characteristic identification module (100) also includes an impedance analysis unit; The impedance analysis unit is used to process the time-domain data using the Fourier transform algorithm to calculate the load-side electrical parameters, which include equivalent resistance, equivalent capacitance, and equivalent inductance. The impedance analysis unit is also used to perform the determination of the load type: If the calculated equivalent resistance is higher than the short-circuit threshold, the value of the equivalent capacitance is not within the measurement noise range, and the value of the equivalent inductance is within the measurement noise range, the load type is determined to be a capacitive load cold start mode, and the equivalent capacitance is retained as a key parameter and the trigger flexible power supply operation is executed. If the calculated equivalent resistance is higher than the short-circuit threshold, and the value of the equivalent capacitance is within the measurement noise range, and the value of the equivalent inductance is within the measurement noise range, the load type is determined to be a standard resistive load mode, and a normal power supply operation is triggered. If the calculated equivalent resistance is less than or equal to the short-circuit threshold, or if the calculated equivalent inductance is not within the measurement noise range, the load type is determined to be either a physical short-circuit mode or a fault mode, and the bidirectional DC-DC converter is locked. The short-circuit threshold is a pre-set safety lower limit value based on the sum of the inherent impedance of the service robot's external power supply cable and the standard contact resistance of the physical interface.

4. A service robot with external emergency power supply capability according to claim 1, characterized in that, The path energy consumption inversion module (200) includes a topology analysis unit; The topology analysis unit is used to periodically retrieve coordinate data and environmental map data from the SLAM navigation system, and plan the optimal return path based on the current coordinate point and the coordinate point of the nearest charging pile in the system database. The topology analysis unit is also used to parse the environmental map data to extract the terrain elevation information of the optimal return path, identify the maximum slope area or high resistance area in the optimal return path by analyzing the elevation change rate, and determine the slope angle of the maximum slope area. The topology analysis unit is also used to calculate the peak driving power required for the service robot to pass through the maximum slope area or high resistance area at full load based on the dynamic model. The specific steps for calculating the peak driving power are as follows: The dynamic model decomposes the driving force of the service robot into a slope component that overcomes gravity and a ground friction resistance. The topology analysis unit calculates the component of gravity that is downward along the slope based on the slope angle as the slope component, calculates the normal force component perpendicular to the slope and combines it with the ground friction coefficient to obtain the ground friction resistance. The sum of the slope component and the ground friction resistance is taken as the total running resistance, and the product of the total running resistance and the preset climbing speed is taken as the driving peak power. The topology analysis unit is also used to calculate the total return distance by performing path integration on the optimal return path.

5. A service robot with external emergency power supply capability according to claim 4, characterized in that, The path energy consumption inversion module (200) also includes a fence calculation unit; The fence calculation unit is used to receive the total return distance and calculate the minimum retention energy according to the retention energy formula; The specific logic for the fence calculation unit to calculate the minimum retention energy is as follows: The product of the total return distance, average driving energy consumption, and safety redundancy coefficient is taken as the dynamic energy part, the product of the basic static power consumption and the return time is taken as the static energy part, and the sum of the dynamic energy part and the static energy part is taken as the minimum reserved energy. The return time is obtained by dividing the total return distance by the average return speed, and the safety redundancy coefficient is set according to the complexity of the application scenario.

6. A service robot with external emergency power supply capability according to claim 1, characterized in that, The thermodynamic verification module (300) includes: The temperature prediction unit is used to collect the current ambient temperature and the real-time temperature of the battery, calculate the predicted temperature value after a preset evaluation window time for continuous power supply at the requested rated power using the battery internal thermal resistance model, and query the temperature power discharge table of the battery management system based on the predicted temperature value to obtain the maximum discharge power. A derating decision unit is used to receive the maximum discharge power, the peak drive power, and the requested rated power, and to calculate the allowable power supply according to the allowable power formula. Through the coordinated operation of the temperature prediction unit and the derating adjudication unit, a correlation constraint between the battery thermal effect and the return trip power performance of the service robot is established. The preset evaluation window time is set based on the duration of the power request.

7. A service robot with external emergency power supply capability according to claim 6, characterized in that, The specific logic by which the derating decision unit calculates the allowable power supply based on the allowable power formula is as follows: If the maximum discharge power is greater than the sum of the power safety margin and the drive peak power, the requested rated power shall be used as the allowable power supply power. If the maximum discharge power is less than or equal to the sum of the power safety margin and the driving peak power, a reverse iterative approximation method is used to monotonically decrease the requested rated power and substitute it into the battery internal thermal resistance model for recalculation until a derating power value is found, such that the new maximum discharge power at the derating power value can be greater than the sum of the power safety margin and the driving peak power, and the derating power value is used as the allowable power supply power. The power safety margin is a power value preset based on the power consumption and aging degradation of the service robot auxiliary system.

8. A service robot with external emergency power supply capability according to claim 1, characterized in that, The flexible power supply control module (400) includes a strategy generation unit; The strategy generation unit is used to respond to the triggering of the flexible power supply operation, receive the device rated voltage and the equivalent capacitance, convert the allowable power supply to the maximum allowable current, calculate the start-up time constant using the equivalent capacitance according to the time constant formula, and generate a target voltage curve that conforms to the exponential growth law using the voltage curve formula based on the start-up time constant and the device rated voltage, so as to execute the soft start.

9. A service robot with external emergency power supply capability according to claim 8, characterized in that, The flexible power supply control module (400) also includes an execution monitoring unit; The execution monitoring unit is used to receive the target voltage curve and input it as a reference signal to the bottom control loop of the bidirectional DC-DC converter, driving the actual output voltage to follow the change of the target voltage curve, thereby providing external power supply; The execution monitoring unit is also used to execute the power cut-off logic during periods of stable power supply: If the remaining battery energy is equal to or lower than the minimum reserve energy, execute the energy cut-off logic; Alternatively, if the battery temperature rise rate exceeds the safety threshold, execute the thermal runaway prevention shutdown logic; Alternatively, if the real-time load current is lower than the no-load threshold, execute the no-load disconnection logic. The safety threshold is a pre-set upper limit of the temperature rise rate based on the chemical characteristics of the battery cell and the critical conditions for thermal runaway, and the no-load threshold is a pre-set lower limit based on the standby current characteristics of the powered device.

10. A control method for a service robot with external emergency power supply capability, characterized in that, Applied to a service robot with external emergency power supply capability as described in any one of claims 1-9 Includes the following steps: The load feature identification module (100) analyzes the electrical parameters of the load side and determines the load type. When the load type is determined to be a capacitive load cold start mode, the equivalent capacitance in the electrical parameters of the load side is extracted and a flexible power supply operation is triggered. The path energy consumption inversion module (200) plans the optimal return path and reverse-engineers the minimum retained energy and peak driving power required for the service robot to safely return home. The thermal dynamics verification module (300) establishes the correlation constraint between the battery thermal effect and the return power performance of the service robot, and calculates the allowable power supply for constraining the power supply process by combining the real-time battery temperature and the peak power of the drive. The flexible power supply control module (400) responds to the triggering of the flexible power supply operation, performs a soft start to supply power to the outside based on the equivalent capacitance and the allowable power supply power, and performs monitoring and power cut-off logic based on the minimum reserved energy during the stable power supply period.

Images