Temperature control heat consumption balancing device and system of humanoid robot

Abstract

The invention relates to the technical field of humanoid robots, in particular to a humanoid robot temperature control heat consumption balancing device and system.The device comprises a humanoid robot, and digital temperature sensors are arranged on a CPU, a battery pack and driving motors at joint positions in the humanoid robot; a trunk and four limbs in the humanoid robot are each provided with a programmable cooling fan for active cooling and cooling fins for passive cooling, and the trunk and the four limbs in the humanoid robot are each provided with an efficient and low-noise air duct. The system comprises the device, an acquisition module, a monitoring module, a planning module and a control module. According to the method, the accuracy of temperature prediction of the humanoid robot parts is improved, and equipment damage caused by abnormal temperature is effectively prevented; temperature data with tendency and seasonal fluctuation can be processed, and the applicability of the model is improved; integration and application in an existing robot system are easy; temperature abnormity can be found in time, and stable operation of the humanoid robot is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of humanoid robots, and in particular to a temperature control and heat consumption balancing device and system for a humanoid robot. Background Art

[0002] In the current humanoid robot system, there are many key devices with high power consumption, including but not limited to the drive motors of each joint, the high-performance CPU responsible for algorithm processing, and the power supply voltage regulator module. Since these devices generate a lot of heat during operation, how to effectively dissipate the heat of these modules has become an urgent problem to be solved when designing compact humanoid robots. It is worth noting that humanoid robot technology is still in its infancy, and the relevant solutions on the market are not yet mature and difficult to meet actual needs.

[0003] Existing humanoid robots do not consider heat dissipation much, and are mostly based on air cooling methods, such as using an electric motor air pump to fill the air inlet with gas, using the rapid flow of gas to achieve heat transfer, thereby reducing the temperature of the robot's components, or adding exhaust fans, installing heat dissipation holes, arranging heat sinks, etc. to achieve robot heat dissipation. These methods are not much different from traditional industrial equipment heat dissipation. It is impossible to achieve thermal energy balance configuration on complex equipment such as humanoid robots.

[0004] The existing technology has the following defects: First, each component dissipates heat separately, and the heat distribution cannot be systematically managed and scheduled, which easily leads to the situation that the temperature of local components is too high, resulting in the overall performance of the product being reduced; second, there is a lack of temperature monitoring capability, and there is no feedback on the heat dissipation effect of each component, which is passive heat dissipation. Its implementation effect cannot be evaluated, and further control cannot be made; third, there is no intelligence: the humanoid robot system lacks overall intelligent thermal energy management capabilities and cannot adapt to the ever-changing working environment and tasks. It is impossible to dynamically adjust the heat consumption distribution according to the work scene and task, and it is impossible to make intelligent decisions according to different scenes. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a temperature control and heat consumption balancing device and system for a humanoid robot.

[0006] The technical problem to be solved by the present invention is achieved by adopting the following technical solutions:

[0007] A temperature control and heat consumption balancing device for a humanoid robot comprises a humanoid robot, wherein a CPU located on a trunk, a battery pack and a driving motor at a joint of the humanoid robot are all provided with digital temperature sensors, the trunk and limbs of the humanoid robot are all provided with programmable cooling fans for active heat dissipation and heat sinks for passive heat dissipation, and the trunk and limbs of the humanoid robot are all provided with high-efficiency and low-noise air ducts.

[0008] As a further improvement of the present invention, the technical parameters of the temperature sensor are: acquisition range: -40°C to 85°C; acquisition error: ±0.3°C within the range of 10°C to 60°C; acquisition accuracy: 0.1°C.

[0009] As a further improvement of the present invention, the technical parameters of the cooling fan are: power supply voltage: 3.3V-5V; signal control mode: PWM; programming interface: RS232 / RS485.

[0010] A temperature control and heat consumption balancing system for a humanoid robot comprises the temperature control and heat consumption balancing device for a humanoid robot, and further comprises:

[0011] The acquisition module is connected to the temperature sensor in the temperature control and heat consumption balancing device, converts the temperature signal into an electrical signal, and processes the collected temperature data. In addition, the acquisition module also collects the operation data of the drive motor in the temperature control and heat consumption balancing device and the usage information of the battery pack;

[0012] The monitoring module is connected to the acquisition module. It constructs a thermal balance map based on the real-time temperature data collected, and develops a user interaction interface. Through the interaction interface, the temperature changes of each component can be viewed and different temperature monitoring strategies and alarm rules can be set.

[0013] The planning module is connected to the monitoring module and includes a data storage system, a time series prediction algorithm module, a strategy library construction, an alarm rule library, and a rule engine. It is responsible for storing temperature data, performing anomaly detection, temperature prediction, generating alarm information, and generating control instructions;

[0014] The control module is connected to the planning module, the cooling fan in the temperature control and heat consumption balancing device, and the heat sink. According to the control instructions generated by the planning module, the working state of the cooling fan is adjusted in real time. In addition, the control module is also connected to the driving motor on the humanoid robot.

[0015] As a further improvement of the present invention, the time series prediction algorithm module adopts an ARIMA model trained for a humanoid robot.

[0016] As a further improvement of the present invention, the process of the ARIMA model is as follows:

[0017] Step 1: Preprocess the temperature data;

[0018] Step 2: Train the ARIMA model based on the historical temperature data on the humanoid robot and adjust the model parameters;

[0019] Step 3: Use the trained model to predict temperature and give estimated temperature values ​​at several future time points;

[0020] Step 4: Compare the predicted results with the actual temperature data and use the anomaly detection algorithm to detect whether there is any anomaly.

[0021] As a further improvement of the present invention, the preprocessing includes differential processing.

[0022] As a further improvement of the present invention, the model parameters include an autoregressive order p, a differential order d, and a moving average order q.

[0023] As a further improvement of the present invention, the anomaly detection algorithm includes hard rules set based on a fixed temperature threshold, soft rules set based on a temperature change trend predicted by a time series, and conditional rules set based on different pre-set trigger conditions.

[0024] The beneficial effects of the present invention are:

[0025] 1) Improved the accuracy of temperature prediction of humanoid robot parts, effectively preventing equipment damage caused by abnormal temperature;

[0026] 2) Through differential processing, the present invention can process temperature data with trend and seasonal fluctuations, thus improving the applicability of the model;

[0027] 3) The ARIMA model method of the present invention is simple and easy to implement and can be easily integrated and applied in existing robot systems;

[0028] 4) Through real-time monitoring and prediction, the present invention can detect temperature anomalies in a timely manner and ensure the stable operation of the humanoid robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention is further described below in conjunction with the accompanying drawings and embodiments:

[0030] Figure 1 It is a schematic diagram of the layout of temperature sensors in the temperature control and heat consumption balancing device;

[0031] Figure 2 This is a schematic diagram of the layout of the cooling fans in the temperature control and heat consumption balancing device;

[0032] Figure 3 This is a system architecture diagram of the present invention;

[0033] Figure 4Flowchart of the time series forecasting algorithm module. DETAILED DESCRIPTION

[0034] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0035] A temperature control and heat consumption balancing device for a humanoid robot, comprising a humanoid robot, such as Figure 1 As shown in the figure, the CPU, battery pack and drive motors at the joints of the humanoid robot are all equipped with digital temperature sensors. The technical parameters of the temperature sensors are as follows: acquisition range: -40℃~85℃; acquisition error: ±0.3℃ within the range of 10℃~60℃; acquisition accuracy: 0.1℃. Figure 2 As shown, the trunk and limbs of the humanoid robot are equipped with programmable cooling fans for active heat dissipation and heat sinks for passive heat dissipation. The technical parameters of the cooling fan are: power supply voltage: 3.3V~5V; signal control mode: PWM; programming interface: RS232 / RS485. The heat sink can accelerate the outflow of heat. At the same time, the trunk and limbs of the humanoid robot are equipped with high-efficiency and low-noise air ducts to achieve convection of air in the cavity under different conditions, taking away the heat of joints and devices.

[0036] like Figure 3 The figure shows the architecture of a temperature control and heat consumption balance system for a humanoid robot. In addition to the temperature control and heat consumption balance device for a humanoid robot, the system also includes four parts: acquisition, monitoring, planning, and control. The details are as follows:

[0037] The acquisition module is connected to the temperature sensor in the temperature control and heat consumption balancing device, converts the temperature signal into an electrical signal, and processes the collected temperature data, including filtering and calibration, to improve the quality and effectiveness of the data. In addition, the acquisition module also collects the operating data of the drive motor in the temperature control and heat consumption balancing device, including current value, position, speed, acceleration, etc., and also collects the usage information of the battery pack, including the battery current, output power, residual power and other information.

[0038] The monitoring module is connected to the acquisition module. It constructs a thermal balance map based on the real-time temperature data collected and provides an API interface for development, which facilitates the development of a user interaction interface. The temperature changes of each component can be viewed through the interactive interface, and different temperature monitoring strategies and alarm rules can be set. The monitoring module can also set personalized temperature monitoring strategies based on seasons, regions, work scenarios and other characteristics, so that users can make more choices and improve the environmental adaptability of the humanoid robot.

[0039] The planning module is connected to the monitoring module, including the data storage system, time series prediction algorithm module, strategy library construction, alarm rule library, and rule engine. It is responsible for storing temperature data, performing anomaly detection, temperature prediction, generating alarm information, and generating control instructions. Among them, the time series prediction algorithm module uses advanced time series prediction algorithms based on the characteristics of temperature data to accurately predict future temperature trends. The algorithm will comprehensively consider historical data, seasonal changes, trends and other factors to improve the accuracy of predictions.

[0040] The time series prediction algorithm is as follows:

[0041] ARIMA model (autoregressive moving average model): Model and predict the temperature data of humanoid robot parts. The core idea is to build an autoregressive model by analyzing past temperature data to predict possible temperature changes in the future. By modeling and analyzing historical temperature data, possible temperature anomalies can be discovered in advance to avoid overheating of the equipment. The key technical points of the ARIMA model are as follows:

[0042] 1) Autoregressive model. Autoregressive (AR) is a statistical model that uses past data points to predict current values. The present invention creatively uses the autoregressive model to process the temperature history data of the humanoid robot and predict subsequent data. The core idea of ​​the autoregressive model is that there is a certain dependency relationship between the current observation value and the historical observation value. By mining this relationship, future values ​​can be predicted.

[0043] The basic principle of the autoregressive model is as follows:

[0044] Suppose there is a set of time series data {Xt}, where t represents time. The autoregressive model assumes that there is a linear relationship between the current observation value Xt and the observation values ​​Xt-1, Xt-2, ..., Xt-p at the past p moments.

[0045] Therefore, we can establish the following mathematical expression:

[0046] Xt=c+φ1Xt-1+φ2Xt-2+…+φpXt-p+εt;

[0047] Among them, c is a constant term, φ1, φ2, …, φp are autoregressive coefficients, and εt is a random error term.

[0048] The autoregressive coefficient reflects the degree of influence of past observations on current observations, while the random error term represents the part that cannot be explained by the model.

[0049] 2) Integration: Eliminate the trend in the data and make the data stable. Differentiation is a commonly used data processing method in time series analysis. Its main purpose is to eliminate the trend in the data and make the originally non-stationary data stable. In the present invention, the stability of the data is an important premise, because non-stationary data often leads to inaccurate model estimation and invalid prediction results. Therefore, differentiation is widely used in data analysis as a simple and effective means of stabilization.

[0050] Non-stationary data usually have the following characteristics: obvious trends, seasonal or cyclical fluctuations over time. These characteristics make the mean, variance and autocovariance of the data change over time, thus violating the assumption of stationary time series. In order to transform non-stationary data into stationary data, the difference method eliminates the long-term trend and seasonal components in the data by subtracting adjacent observations.

[0051] Moving Average (MA): Improves forecast accuracy by modeling the change in forecast error. The present invention improves forecast accuracy by modeling the change in forecast error. The basic idea of ​​the present invention is to smooth time series data by taking a weighted average of observations over a certain period of time, thereby revealing potential trends or patterns and reducing the impact of random fluctuations on forecast results.

[0052] The process of ARIMA model is as follows Figure 4 As shown, the details are as follows:

[0053] Step 1: Preprocess the temperature data, including differencing to remove trends and seasonal fluctuations in the data.

[0054] Step 2: Train the ARIMA model based on the historical temperature data on the humanoid robot and adjust the model parameters, including the autoregressive order p, the difference order d, and the moving average order q.

[0055] Step 3: Use the trained model to predict temperature and give estimated temperature values ​​at several future time points

[0056] Step 4: Compare the predicted results with the actual temperature data and use the anomaly detection algorithm to detect whether there is any anomaly.

[0057] The anomaly detection algorithm combines hard rules, soft rules and conditional rules to dynamically monitor and warn the equipment on the humanoid robot. Among them, hard rules: set a fixed temperature threshold, and immediately alarm when the component temperature exceeds the threshold. Soft rules: Based on the temperature change trend predicted by the time series, an early warning is issued when it is detected that the safe temperature range is about to be exceeded. Conditional rules: Different trigger conditions are set in advance according to different conditions such as seasons and environments, and the rules are triggered when the conditions are met.

[0058] The code example is as follows:

[0059] #Preprocess the data by difference

[0060] diff_temp = diff(temp);

[0061] #Use ARIMA model for temperature prediction

[0062] model = arima(2,1,2); #Assuming 2nd order autoregression, 1st order difference, 2nd order moving average

[0063] fit_model=estimate(model,diff_temp);

[0064] #Predict future temperatures

[0065] num_steps = 10; #Predict the next 10 time points

[0066] forecast=forecast(fit_model,num_steps);

[0067] #Restore the predicted value to the original scale

[0068] forecast_temp=temp(end)+cumsum(forecast);

[0069] #Set hard rule temperature threshold

[0070] threshold=75; #temperature threshold

[0071] if any(forecast_temp>threshold)

[0072] disp('Warning: The predicted temperature exceeds the specified threshold, the device needs to be checked!');

[0073] end

[0074] #Set conditional rule trigger

[0075] if any(trigger(forecast_temp)>0)

[0076] disp('Warning: The predicted temperature exceeds the set seasonal threshold, check the device!');

[0077] End

[0078] This embodiment provides a temperature control and heat consumption balancing device and system for a humanoid robot, which is highly advanced. It takes a leading position in the industry by creatively using the ARIMA model in the field of humanoid robots and establishing a model that uses the data of humanoid robots as training data.

[0079] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A temperature control and heat consumption balancing device for a humanoid robot, comprising a humanoid robot, characterized in that: The CPU on the torso, the battery pack and the drive motors at the joints of the humanoid robot are all equipped with digital temperature sensors. The torso and limbs of the humanoid robot are all equipped with programmable cooling fans for active heat dissipation and heat sinks for passive heat dissipation. The torso and limbs of the humanoid robot are all equipped with high-efficiency and low-noise air ducts.

2. The temperature control and heat consumption balancing device of a humanoid robot according to claim 1, characterized in that: The technical parameters of the temperature sensor are: acquisition range: -40℃~85℃; acquisition error: ±0.3℃ in the range of 10℃~60℃; acquisition accuracy: 0.1℃.

3. The temperature control and heat consumption balancing device of a humanoid robot according to claim 1, characterized in that: The technical parameters of the cooling fan are: power supply voltage: 3.3V~5V; signal control mode: PWM; programming interface: RS232 / RS485.

4. A temperature control and heat consumption balancing system for a humanoid robot, characterized in that: A temperature control and heat consumption balancing device for a humanoid robot comprising any one of claims 1 to 3, further comprising: The acquisition module is connected to the temperature sensor in the temperature control and heat consumption balancing device, converts the temperature signal into an electrical signal, and processes the collected temperature data. In addition, the acquisition module also collects the operation data of the drive motor in the temperature control and heat consumption balancing device and the usage information of the battery pack; The monitoring module is connected to the acquisition module. It constructs a thermal balance map based on the real-time temperature data collected, and develops a user interaction interface. Through the interaction interface, the temperature changes of each component can be viewed and different temperature monitoring strategies and alarm rules can be set. The planning module is connected to the monitoring module and includes a data storage system, a time series prediction algorithm module, a strategy library construction, an alarm rule library, and a rule engine. It is responsible for storing temperature data, performing anomaly detection, temperature prediction, generating alarm information, and generating control instructions; The control module is connected to the planning module, the cooling fan in the temperature control and heat consumption balancing device, and the heat sink. According to the control instructions generated by the planning module, the working state of the cooling fan is adjusted in real time. In addition, the control module is also connected to the driving motor on the humanoid robot.

5. The temperature control and heat consumption balancing system of a humanoid robot according to claim 4, characterized in that: The time series prediction algorithm module uses the ARIMA model trained for humanoid robots.

6. The temperature control and heat consumption balancing system of a humanoid robot according to claim 5, characterized in that: The process of the ARIMA model is as follows: Step 1: Preprocess the temperature data; Step 2: Train the ARIMA model based on the historical temperature data on the humanoid robot and adjust the model parameters; Step 3: Use the trained model to predict temperature and give estimated temperature values ​​at several future time points; Step 4: Compare the predicted results with the actual temperature data and use the anomaly detection algorithm to detect whether there is any anomaly.

7. The temperature control and heat consumption balancing system of a humanoid robot according to claim 6, characterized in that: Preprocessing includes differential processing.

8. The temperature control and heat consumption balancing system of a humanoid robot according to claim 6, characterized in that: The model parameters include the autoregressive order p, the difference order d, and the moving average order q.

9. The temperature control and heat consumption balancing system of a humanoid robot according to claim 6, characterized in that: The anomaly detection algorithm includes hard rules set based on fixed temperature thresholds, soft rules set based on temperature change trends predicted by time series, and conditional rules set based on pre-set different trigger conditions.

Images