Automatic charging method for robot dog

Abstract

The invention discloses a robot dog automatic charging method, and relates to the technical field of intelligent robots, and the method comprises the steps: installing a sensor in a robot dog, monitoring the electric quantity of a battery, collecting environment data, carrying out the preprocessing, constructing a future electric quantity consumption prediction model, inputting the current electric quantity, the environment data and task demands into the future electric quantity consumption prediction model, and carrying out the automatic charging of the robot dog. Predicting future power consumption; setting a charging threshold value according to future power consumption, and determining a charging strategy; the robot dog determines the position of the charging station according to the charging strategy, returns to the charging station to be butted with the charging interface, and selects a charging mode; the robot dog starts to be charged, meanwhile, built-in heat dissipation management equipment is started, and the battery state is monitored in real time. According to the invention, the future power consumption is predicted by using the future power consumption prediction model, high-precision power prediction is realized, and the robot dog can dynamically adjust the charging strategy in a complex environment, so that the success rate and efficiency of task completion are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent robots, and in particular to an automatic charging method for a robot dog. Background Art

[0002] In recent years, with the advancement of robotics and artificial intelligence, robot dogs have been widely used in patrol, search and rescue, and logistics. Despite their high levels of autonomy and intelligence, battery life remains a major constraint to their application. To address this issue, automated charging technology has become a research hotspot. Early approaches relied on fixed charging stations and docking and charging via wireless communication or visual recognition, but these methods still have numerous shortcomings.

[0003] With advances in sensor technology, modern robot dogs are equipped with high-precision power and environmental sensors to monitor battery status and environmental conditions in real time. Machine learning-based algorithms have improved the accuracy of power predictions, enabling the robot dogs to dynamically adjust their charging strategies. However, existing methods still face challenges, such as the difficulty in accurately predicting power consumption and developing appropriate charging strategies under changing mission demands. In addition, insufficient thermal management during fast charging can shorten battery life and pose safety risks. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for automatically charging a robot dog to solve the problems of inaccurate power prediction and insufficient heat management during fast charging.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for automatically charging a robot dog, which comprises:

[0008] Sensors are installed inside the robot dog to monitor the battery level, collect environmental data, and perform preprocessing to build a future power consumption prediction model. The current power level, environmental data, and task requirements are input into the future power consumption prediction model to predict future power consumption.

[0009] Set charging thresholds based on future power consumption and determine charging strategies;

[0010] The robot dog determines the location of the charging station based on the charging strategy, returns to the charging station, docks with the charging port, and selects the charging mode;

[0011] The robot dog starts charging, activates the built-in heat management device, and monitors the battery status in real time;

[0012] Set charging rate adjustment rules and adjust the charging rate. Generate a charging report after charging is completed, and use the charging report to optimize the model parameters of the future power consumption prediction model.

[0013] As a preferred solution of the automatic charging method for the robot dog of the present invention, wherein: a sensor is installed inside the robot dog to monitor the battery power, collect environmental data, and perform preprocessing, the specific steps are as follows:

[0014] The sensors are power sensors, temperature sensors and humidity sensors;

[0015] Use a power sensor to monitor the battery power, and use a temperature sensor and humidity sensor to collect temperature and humidity values;

[0016] Normalize the battery power, temperature, humidity and mission requirements.

[0017] As a preferred solution of the automatic charging method for the robot dog of the present invention, the following steps are used to construct a future power consumption prediction model, input the current power consumption, environmental data and task requirements into the future power consumption prediction model, and predict future power consumption:

[0018] Select a combination of reinforcement learning and long short-term memory network algorithms to build a future power consumption prediction model;

[0019] Set up the architecture of the future power consumption prediction model based on reinforcement learning and long short-term memory network;

[0020] The architecture of the future power consumption prediction model includes an input layer, an LSTM layer, a fully connected layer, a reinforcement learning module, and an output layer;

[0021] Combine the current battery power, temperature, humidity, and task requirements into a power consumption dataset;

[0022] The power consumption dataset is divided into a training set, a test set, and a validation set, which are input into the future power consumption prediction model for training, testing, and validation respectively;

[0023] The current battery power, temperature, humidity, and task requirements are input into a future power consumption prediction model to predict future power consumption.

[0024] As a preferred solution of the automatic charging method for the robot dog of the present invention, wherein: the charging threshold is set according to the future power consumption, and the charging strategy is determined, the specific steps are:

[0025] Preset charging thresholds based on environmental data and mission requirements, and adjust charging thresholds based on future power consumption;

[0026] The charging strategy refers to the selected charging mode and charging station location;

[0027] Use the Haversine formula to calculate the distance between the robot dog's location and each charging station location, screen the available charging stations, and select the optimal charging station location.

[0028] As a preferred embodiment of the automatic charging method for a robot dog according to the present invention, the robot dog determines the location of a charging station according to a charging strategy, returns to the charging station to dock with a charging interface, and selects a charging mode. The specific steps are as follows:

[0029] Use the improved A* algorithm to calculate the shortest path between the robot dog's location and the charging station's location;

[0030] Define the starting point as the robot dog's location and the target point as the charging station's location. Initialize the open list and closed list, add the starting point to the open list, select the target point from the open list for path planning, and the robot dog moves along the planned path. When the robot dog arrives at the charging station, it identifies the charging interface type, connects to the charging interface based on the identification result, and selects the charging mode according to the charging strategy.

[0031] As a preferred embodiment of the automatic charging method for the robot dog of the present invention, the robot dog starts charging, simultaneously activates the built-in heat dissipation management device, and monitors the battery status in real time. The specific steps are as follows:

[0032] Activate the robot dog's built-in heat management device to adjust the fan speed according to the battery temperature;

[0033] The battery status refers to battery temperature and charging current.

[0034] As a preferred solution of the automatic charging method for the robot dog of the present invention, wherein: the setting of the charging rate adjustment rule and adjusting the charging rate, and generating a charging report after charging is completed, the specific steps are as follows:

[0035] Set charging rate adjustment rules based on battery temperature, charging current and current battery charge;

[0036] The charging rate is adjusted based on the charging rate adjustment rule, and the charging time, battery power change and charging rate are recorded as a charging report.

[0037] As a preferred embodiment of the automatic charging method for a robot dog according to the present invention, the method of using the charging report to optimize the model parameters of the future power consumption prediction model comprises the following specific steps:

[0038] Standardize charging reports using standardized templates;

[0039] The charging report and power consumption dataset are combined into a comprehensive dataset to optimize the model parameters of the future power consumption prediction model.

[0040] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the robot dog automatic charging method as described in the first aspect of the present invention is implemented.

[0041] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the robot dog automatic charging method as described in the first aspect of the present invention is implemented.

[0042] The present invention achieves the following beneficial effects: By using a future power consumption prediction model to predict future power consumption, high-precision power forecasting is achieved, enabling the robot dog to dynamically adjust its charging strategy in complex environments, avoiding low power situations and thus improving the success rate and efficiency of task completion. By activating the built-in heat management device and monitoring the battery status in real time, effective heat management and dynamic charging rate adjustment are achieved, ensuring the safety and efficiency of the charging process, extending the battery life, and reducing maintenance costs and safety risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a flow chart of the automatic charging method for the robot dog in Example 1.

[0045] Figure 2 This is a schematic diagram of setting a charging threshold according to future power consumption in Example 1. DETAILED DESCRIPTION

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0047] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0048] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0049] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a method for automatically charging a robot dog, comprising the following steps:

[0050] S1: Install sensors inside the robot dog to monitor the battery level, collect environmental data, and perform preprocessing.

[0051] The specific steps are:

[0052] S1.1. Sensors refer to power sensors, temperature sensors, and humidity sensors;

[0053] The power sensor uses the high-precision coulomb counter sensor MAX17055, the temperature sensor uses the digital temperature sensor DS18B20, and the humidity sensor uses the capacitive humidity sensor HIH6130.

[0054] It should also be noted that these sensors were selected based on their high precision and stability to ensure the accuracy of data collection.

[0055] S1.2. Use the power sensor to monitor the battery power and record the current battery power once a minute. Use the temperature sensor and humidity sensor to collect the temperature and humidity values once a minute.

[0056] The collected data is transmitted to the cloud server via MQTT through the wireless communication module;

[0057] The current battery power, temperature, humidity, and task requirements are normalized using minimum and maximum normalization.

[0058] It should also be noted that the MQTT protocol can ensure low latency and high reliability of data transmission, making it suitable for real-time monitoring applications. In addition, to improve the security of data transmission, TLS encryption technology can be used. Data is collected once a minute to ensure real-time and continuity of the data and avoid information loss due to long sampling intervals.

[0059] S2: Build a future power consumption prediction model, input the current power consumption, environmental data and task requirements into the future power consumption prediction model, and predict future power consumption.

[0060] The specific steps are:

[0061] S2.1. Select a combination of reinforcement learning and long short-term memory (LSTM) to build a future power consumption prediction model. This combination uses LSTM to capture long-term dependencies in time series and optimizes the model parameters based on actual feedback through the reinforcement learning module.

[0062] Set up the architecture of the future power consumption prediction model based on reinforcement learning and long short-term memory network;

[0063] The architecture of the future power consumption prediction model refers to the input layer, LSTM layer, fully connected layer, reinforcement learning module and output layer;

[0064] Input layer: Receives data such as the current battery charge, temperature, humidity, and task requirements.

[0065] LSTM layer: used to process time series data and capture long-term dependencies between different time points.

[0066] Fully Connected Layer: Maps the output of the LSTM layer to the final prediction space to generate a normalized value of future power consumption.

[0067] Reinforcement learning module: adjusts model parameters based on actual feedback to optimize prediction results.

[0068] Output layer: Generates the final predicted value of future power consumption.

[0069] Combine the current battery power, temperature, humidity, and task requirements into a power consumption dataset;

[0070] The power consumption dataset is divided into a training set, a test set, and a validation set, which are input into the future power consumption prediction model for training, testing, and validation respectively;

[0071] The model is trained using the training set and the model parameters are updated through the back-propagation algorithm.

[0072] A loss function (such as mean squared error (MSE)) is used to measure the difference between the predicted value and the actual value, and the model parameters are adjusted based on this difference.

[0073] Use the test set to evaluate the performance of the model, record the prediction error, and adjust the model parameters based on the error.

[0074] Use the validation set to further fine-tune the model to ensure its stability and accuracy on new data.

[0075] The current battery power, temperature, humidity, and task requirements are input into the future power consumption prediction model to predict the future power consumption. The expression for the predicted power consumption is:

[0076] y=f(h t―1 ,[E,T,H,D],X t―1 ,θ);

[0077] Among them, y is the normalized value of future power consumption, f is the future power consumption prediction model, and h t―1 is the output of the LSTM layer at time step t-1, E is the current battery charge, T is the temperature value, H is the humidity value, D is the task requirement, which refers to the energy consumption requirement of the current task, X t―1 is the memory unit of time step t-1, and θ is the parameter of the future power consumption prediction model.

[0078] S3: Set a charging threshold based on future power consumption and determine a charging strategy.

[0079] The specific steps are:

[0080] S3.1. Based on environmental data and mission requirements, the battery's minimum safe charge level is preset to 50%. The charge threshold is adjusted based on the ratio of future power consumption to current power.

[0081] If the predicted future power consumption exceeds 70% of the current power, the charging threshold is raised to 60%, causing the robot dog to trigger the charging demand earlier. If the predicted future power consumption is less than 30% of the current power, the charging threshold is lowered to 40%, delaying the charging demand.

[0082] In high temperature or high humidity environments, the charging threshold can be appropriately increased, and in low temperature or dry environments, the charging threshold can be appropriately lowered.

[0083] It should also be noted that dynamically adjusting the charging threshold can better adapt to different environments and task requirements, ensuring that the robot dog returns to the charging station in time when the battery is low.

[0084] S3.2, Charging strategy refers to the selected charging mode and charging station location;

[0085] Determine whether charging is needed based on the current power level and future power consumption. If the current power level is lower than the charging threshold or the future power consumption is greater than the current power level, the charging requirement is triggered, otherwise the task continues.

[0086] Select the charging mode according to the urgency of the task. If the task is more urgent, select the fast charging mode; otherwise, select the normal charging mode.

[0087] It's also important to note that in practical applications, historical data analysis can be used to determine task urgency standards. For example, if a task needs to be completed quickly and is time-sensitive, it's considered high urgency; otherwise, it's considered low urgency. This allows for the appropriate selection of fast or standard charging modes based on task urgency, ensuring that task completion is prioritized in critical situations.

[0088] S3.3. Obtain the current location of the robot dog through GPS, use the Haversine formula to calculate the distance between the robot dog's location and the location of each charging station, filter the available charging stations according to the charging mode, and select the optimal charging station location.

[0089] The range of charging stations that the robot dog can reach is calculated based on the current power of the robot dog and the power consumption per unit distance.

[0090] The charging stations that meet the required charging mode are selected within the range of charging stations that the robot dog can reach.

[0091] From the filtered available charging stations, the nearest charging station is selected as the optimal charging station, and the location of the charging station is recorded.

[0092] It should also be noted that the Haversine formula is used to calculate the shortest distance between two points on the Earth's surface. For example, if the robot dog's current location is 40.7128°N, 74.0060°W, and a charging station is located at 40.7127°N, 74.0061°W, the Haversine formula can accurately calculate the distance between the two, facilitating subsequent route planning. If the nearest charging station is unavailable (for example, occupied or faulty), the next closest station is selected to ensure continuous and reliable charging.

[0093] S4: The robot dog determines the location of the charging station according to the charging strategy, returns to the charging station, docks with the charging interface, and selects the charging mode.

[0094] The specific steps are:

[0095] S4.1. Calculate the shortest path between the robot dog's location and the charging station's location using the improved A* algorithm;

[0096] In the traditional A* algorithm, the weight of the heuristic estimated cost from the node to the target point is fixed and cannot be adjusted dynamically, resulting in low efficiency in some scenarios.

[0097] The improved A* algorithm introduces dynamic weights and dynamically adjusts the weights according to the complexity of the environment and the power status. The improved A* algorithm expression is:

[0098] c(n)=g(n)+α·h(n);

[0099] Among them, c(n) is the shortest path between the robot dog's location and the charging station's location, g(n) is the actual cost from the starting point to node n, α is the dynamic weight of the heuristic estimated cost from node n to the target point, and h(n) is the heuristic estimated cost from node n to the target point.

[0100] The dynamic weight of the heuristic estimated cost from node n to the target point is calculated as:

[0101]

[0102] Among them, C is the environmental complexity and B is the maximum environmental complexity.

[0103] It should also be noted that the improved A algorithm introduces a dynamic weight adjustment mechanism, which enables the heuristic estimation cost to be dynamically adjusted according to actual conditions. For example, in complex environments (such as areas with dense obstacles), increasing the weight makes the heuristic estimation cost larger, thereby reducing the search range and improving search efficiency. The dynamic weight adjustment mechanism enables the A* algorithm to better adapt to complex environments and improve the efficiency and flexibility of path planning.

[0104] S4.2. Define the starting point as the robot dog's location and the target point as the charging station's location. Initialize the open list (OpenList) and the closed list (ClosedList). The open list stores the nodes to be explored, and the closed list stores the nodes that have been explored.

[0105] Add the starting point to the open list (to prevent missing the case where the starting point is also the target point), set its g(n) = 0, and calculate the value of c(n).

[0106] Select the node with the smallest c(n) value from the open list as the current node. If the current node is the target point, the path planning is completed and the path is backtracked. Otherwise, move the current node from the open list to the closed list.

[0107] The backtracking path refers to tracing back from the target point to the starting point, recording all the nodes on the path, and forming a planned path.

[0108] The robot dog moves along a planned path. When it reaches a charging station, it uses a camera and infrared sensor to identify the charging port type. Based on the identification result, it adjusts the robotic arm to dock with the charging port. Once docked, it selects a charging mode based on the charging strategy. It then interacts with the charging station through a communication module to confirm the maximum charging power supported by both parties.

[0109] It's also worth noting that the camera captures images of the charging port and uses image recognition algorithms to determine the port type (e.g., Type-C, Micro-USB, etc.). Infrared sensors assist in detecting the port's position, ensuring accurate docking by the robotic arm. This multi-sensor fusion approach improves the accuracy and reliability of charging port recognition, enabling efficient charging docking.

[0110] S5: The robot dog starts charging, activates the built-in heat management device, and monitors the battery status in real time.

[0111] The specific steps are:

[0112] S5.1. Activate the robot's built-in thermal management device, including the cooling fan and heat sink, and adjust the fan speed according to the battery temperature.

[0113] Monitor the battery temperature in real time. When the battery temperature is lower than the optimal operating temperature, the fan runs at a low speed to save energy. When the battery temperature is higher than the optimal operating temperature, the fan speed is increased.

[0114] It should also be noted that the startup and adjustment of the heat management device are based on the real-time monitoring of the battery temperature. The specific control logic can be linear adjustment or segmented adjustment, which ensures the heat dissipation effect while saving energy consumption and improving the safety and stability of the battery during charging.

[0115] S5.2. Battery status refers to battery temperature and charging current.

[0116] Monitor battery temperature and charging current in real time to ensure that the current value is within the maximum charging current range allowed by the battery. When the charging current is close to the maximum charging current or the battery temperature is too high, reduce the current to prevent overload.

[0117] It should also be noted that real-time monitoring and adjustment of battery status can effectively prevent overheating and overcurrent, and extend battery life.

[0118] S6: Set charging rate adjustment rules and adjust the charging rate, and generate a charging report after charging is completed.

[0119] The specific steps are:

[0120] S6.1. Set charging rate adjustment rules based on battery temperature, charging current, and current battery charge;

[0121] When the temperature reaches the battery temperature threshold, the charging rate is reduced to prevent the battery from overheating.

[0122] For example, when the battery temperature reaches 40°C, the charging rate is reduced to 80% of the original rate.

[0123] When the battery's current charge reaches 90%, the charging rate is gradually reduced to ensure that the battery is fully charged safely.

[0124] For example, when the battery charge reaches 90%, the charging rate is gradually reduced to 50% of the original rate.

[0125] When the charging current reaches the maximum charging current allowed by the battery, the charging rate is reduced to prevent overcurrent.

[0126] When the temperature is close to the optimal operating temperature of the battery, the current battery charge is low, and the charging current is close to the optimal charging current of the battery, the charging rate is increased to the maximum power supported by the charging station to achieve fast charging.

[0127] It should also be noted that the charging rate adjustment rules take into account the battery temperature, charging current and current charge to ensure the safety and efficiency of the charging process.

[0128] S6.2. Adjust the charging rate based on the charging rate adjustment rule, and record the charging time, battery charge change, and charging rate as a charging report.

[0129] It should also be noted that: charging time is the duration of the charge, battery charge change is the increase in battery charge at the end of charging compared to the beginning, and charging rate is the average charging rate of the charge. Generating a charging report helps with subsequent analysis and optimization of charging strategies, ensuring the transparency and controllability of the charging process. The charging report format can be JSON or CSV file to facilitate subsequent data processing and analysis.

[0130] S7: Optimize model parameters of a future power consumption prediction model using the charging report.

[0131] The specific steps are:

[0132] S7.1. Standardize charging reports using standardized templates to ensure data are in the same dimension.

[0133] The charging time and battery capacity change are standardized using minimum and maximum normalization, and the charging rate is standardized using Z-score.

[0134] It should also be noted that the standardized template can specify that the charging start time is a UTC timestamp, the end time is a UTC timestamp, the power change is a percentage, and the average charging rate is the percentage of power increase per minute. This unified format facilitates data integration and analysis of different devices. The standardized charging report facilitates subsequent data analysis and improves data consistency and comparability.

[0135] S7.2. Combine the charging report and power consumption datasets into a comprehensive dataset. Select the mean square error as the loss function. Use the comprehensive dataset to perform gradient descent optimization on the model parameters of the future power consumption prediction model. Calculate the gradient using the backpropagation algorithm and update the model parameters of the future power consumption prediction model. The expression is:

[0136] θ′=θ―η·M;

[0137] Among them, θ′ is the model parameter of the updated future power consumption prediction model, η is the learning rate, and M is the gradient of the loss function with respect to the parameters of the future power consumption prediction model.

[0138] This embodiment also provides a computer device suitable for the automatic charging method of a robot dog, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the automatic charging method of the robot dog proposed in the above embodiment.

[0139] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0140] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing automatic charging of a robot dog as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0141] In summary, the present invention achieves high-precision power prediction by using a future power consumption prediction model to predict future power consumption. This allows the robot dog to dynamically adjust its charging strategy in complex environments, avoiding low-power situations and thus improving the success rate and efficiency of task completion. By activating the built-in heat management device and monitoring the battery status in real time, effective heat management and dynamic charging rate adjustment are achieved, ensuring the safety and efficiency of the charging process, extending the battery life, and reducing maintenance costs and safety risks.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for automatically charging a robot dog, characterized by: include, Sensors are installed inside the robot dog to monitor the battery level, collect environmental data, and perform preprocessing to build a future power consumption prediction model. The current power level, environmental data, and task requirements are input into the future power consumption prediction model to predict future power consumption. Set charging thresholds based on future power consumption and determine charging strategies; The robot dog determines the location of the charging station based on the charging strategy, returns to the charging station, docks with the charging port, and selects the charging mode; The robot dog starts charging, activates the built-in heat management device, and monitors the battery status in real time; Set charging rate adjustment rules and adjust the charging rate. Generate a charging report after charging is completed, and use the charging report to optimize the model parameters of the future power consumption prediction model.

2. The automatic charging method for a robot dog according to claim 1, wherein: The sensor is installed inside the robot dog to monitor the battery power, collect environmental data, and perform preprocessing. The specific steps are as follows: The sensors are power sensors, temperature sensors and humidity sensors; Use a power sensor to monitor the battery power, and use a temperature sensor and humidity sensor to collect temperature and humidity values; Normalize the current battery power, temperature, humidity, and mission requirements.

3. The automatic charging method for a robot dog according to claim 2, wherein: The future power consumption prediction model is constructed by inputting the current power consumption, environmental data and task requirements into the future power consumption prediction model to predict future power consumption. The specific steps are as follows: Select a combination of reinforcement learning and long short-term memory network algorithms to build a future power consumption prediction model; Set up the architecture of the future power consumption prediction model based on reinforcement learning and long short-term memory network; The architecture of the future power consumption prediction model includes an input layer, an LSTM layer, a fully connected layer, a reinforcement learning module, and an output layer; Combine the current battery power, temperature, humidity, and task requirements into a power consumption dataset; The power consumption dataset is divided into a training set, a test set, and a validation set, which are input into the future power consumption prediction model for training, testing, and validation respectively; The current battery power, temperature, humidity, and task requirements are input into a future power consumption prediction model to predict future power consumption.

4. The automatic charging method for a robot dog according to claim 3, wherein: The specific steps of setting the charging threshold according to future power consumption and determining the charging strategy are as follows: Preset charging thresholds based on environmental data and mission requirements, and adjust charging thresholds based on future power consumption; The charging strategy refers to the selected charging mode and charging station location; Use the Haversine formula to calculate the distance between the robot dog's location and each charging station location, screen the available charging stations, and select the optimal charging station location.

5. The automatic charging method for a robot dog according to claim 4, characterized in that: The robot dog determines the location of the charging station according to the charging strategy, returns to the charging station to dock with the charging interface, and selects the charging mode. The specific steps are as follows: Use the improved A* algorithm to calculate the shortest path between the robot dog's location and the charging station's location; Define the starting point as the robot dog's location and the target point as the charging station's location. Initialize the open list and closed list, add the starting point to the open list, select the target point from the open list for path planning, and the robot dog moves along the planned path. When the robot dog arrives at the charging station, it identifies the charging interface type, connects to the charging interface based on the identification result, and selects the charging mode according to the charging strategy.

6. The automatic charging method for a robot dog according to claim 5, characterized in that: The robot dog starts charging, starts the built-in heat dissipation management device, and monitors the battery status in real time. The specific steps are as follows: Activate the robot dog's built-in heat management device to adjust the fan speed according to the battery temperature; The battery status refers to battery temperature and charging current.

7. The automatic charging method for a robot dog according to claim 6, wherein: The steps of setting the charging rate adjustment rule, adjusting the charging rate, and generating a charging report after charging is completed are as follows: Set charging rate adjustment rules based on battery temperature, charging current and current battery charge; The charging rate is adjusted based on the charging rate adjustment rule, and the charging time, battery power change and charging rate are recorded as a charging report.

8. The automatic charging method for a robot dog according to claim 7, wherein: The specific steps of using the charging report to optimize the model parameters of the future power consumption prediction model are as follows: Standardize charging reports using standardized templates; The charging report and power consumption dataset are combined into a comprehensive dataset to optimize the model parameters of the future power consumption prediction model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the automatic charging method for a robot dog according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the automatic charging method for a robot dog according to any one of claims 1 to 8 are implemented.

Images