Auxiliary monitoring and full-stop decision-making method for robot charging

Abstract

The present invention discloses an auxiliary monitoring and full-time automatic stop decision method for robot charging. According to the auxiliary monitoring and full-time automatic stop decision method for robot charging, after the robot to be charged sends a charging request, the charging system collects and analyzes the initial data of the robot's battery, provides a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging, and then starts charging. During the charging process, the charging terminal monitors the charging voltage, current and other parameters in real time. When the estimated charging time is reached, the charging terminal compares the battery calibrated parameters with the real-time parameters to determine whether the battery is full, and makes a self-stop decision after it is full, thereby avoiding damage to the robot battery due to overcharging. The auxiliary monitoring and full-time automatic stop decision method for robot charging performs statistical analysis and records the charging data of different robots during the charging process, optimizes the robot charging self-stop decision plan through autonomous learning, and enables the robot to use the best charging self-stop plan each time it is charged.

Description

Technical Field

[0001] The present invention relates to the technical field of program design and writing, and in particular to an auxiliary monitoring method for robot charging and a full-fill automatic stop decision method. Background Art

[0002] In our daily lives, we have produced many intelligent robots to help us live, work, and study. However, the operation of intelligent robots cannot do without electricity. Generally, robots are designed with built-in rechargeable batteries. Robots with built-in rechargeable batteries have a large activity area and are not restricted by cables. However, when the robot runs out of power after a period of operation, the battery needs to be recharged. Currently, many robots are charged manually. The battery is charged by connecting the charging power supply and stopped by unplugging the charging power supply. However, when charging the robot, people reserve a long time for the robot to be fully charged. This causes many robots' batteries to remain in the charging state for a long time. The long-term charging state of the robot can easily cause heating of the circuit and damage to the rechargeable battery. In addition, since robots experience many charging and discharging events in their life cycle, flexible switching between different charging states is also crucial. A gentle current switching mechanism can extend the battery life. To this end, the applicant has designed and compiled a robot charging auxiliary monitoring and full-time automatic stop decision-making method based on the needs of robot charging. Through the robot charging auxiliary monitoring and full-time automatic stop decision-making method, the robot battery is automatically stopped after being fully charged, avoiding the robot battery being in the charging state for a long time, which may cause battery damage and safety accidents. Summary of the Invention

[0003] In response to the above-mentioned shortcomings, the present invention provides an auxiliary monitoring and full-time automatic stop decision method for robot charging. After the charging cable is connected and the robot to be charged sends a charging request, the charging system collects and analyzes the initial battery data of the robot, provides a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging, and then starts charging. During the charging process, the charging terminal monitors the charging voltage, current and other parameters in real time. When the estimated charging time is reached, the charging terminal compares the battery calibrated parameters with the real-time parameters to determine whether the battery is full, and makes a self-stop decision after it is full, thereby avoiding damage to the robot battery due to overcharging; the auxiliary monitoring and full-time automatic stop decision method for robot charging performs statistical analysis and records the charging data of different robots during the charging process, optimizes the robot charging self-stop decision plan through autonomous learning, and enables the robot to use the best charging self-stop plan each time it is charged.

[0004] The present invention is achieved through the following technical solutions: a method for auxiliary monitoring and full-time automatic stop decision-making for robot charging, characterized in that: after the robot to be charged issues a charging request, the charging system collects and analyzes the initial battery data of the robot, provides a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging, and then starts charging. The charging process involves three processes: a pre-charging stage, in which the charging voltage and current are detected to determine whether they meet the robot's charging requirements. If they do not meet the requirements, the robot's charging battery information is re-collected, and then a charging plan and time are re-given. After the pre-charging stage detects that the charging requirements are met, the pre-charging time is waited for and then the fast charging stage is entered. After the fast charging stage time is reached, the floating charging stage is entered. After the floating charging stage time is reached, the charging system terminal collects the robot's battery information, and then compares it with the battery calibration parameters to determine whether the battery is full. After it is full, an automatic stop decision is made to avoid damage to the robot battery due to overcharging. During the charging process, the charging terminal monitors the charging voltage, current and other parameters in real time to avoid battery damage due to abnormal voltage and current.

[0005] As a preferred technical solution of the present invention, the robot charging auxiliary monitoring and full-time automatic stop decision method performs statistical analysis and records on the charging data of different robots during the charging process, and optimizes the robot charging automatic stop decision plan through autonomous learning, so that the robot uses the best charging automatic stop plan each time it charges.

[0006] As an optimal technical solution of the present invention, the robot charging auxiliary monitoring and full-time automatic stop decision method collects the robot battery information and estimates the time for full charging. The charging time includes the pre-charging stage time, the fast charging stage time and the floating charging stage time.

[0007] As a preferred technical solution of the present invention, the described robot charging auxiliary monitoring and full-charge automatic stop decision method has a smooth charging state switching mechanism: at the end of the pre-charge phase, a prediction model is used to predict the remaining time to reach the specified starting SOC of the rapid charge phase. If the predicted time is less than the pre-charge-to-rapid charge switching threshold, the charging current is adjusted to half the sum of the stable currents of the pre-charge and rapid charge phases. At the end of the rapid charge phase, a prediction model is used to predict the remaining time to reach the specified full-charge SOC. If the predicted time is less than the rapid charge-to-full charge switching threshold, the charging current is adjusted to half the stable current of the rapid charge phase.

[0008] The charging state soft switching mechanism prediction model is a remaining charging time prediction model based on the attention mechanism to improve the BP algorithm. The specific prediction method is: first, the battery information including voltage, current, SOC, etc. in the past time window is extracted into a feature vector X:

[0009] X=x1,x2,x3,…xn

[0010] In the formula, n is the length of the vector. The attention mechanism is constructed to improve the BP algorithm, which consists of an attention mechanism layer and a fully connected layer with n input nodes and 1 output node. The calculation method of the attention mechanism layer is

[0011] Xa=σaWaX+ba

[0012] Ta=softmaxXa

[0013] Where Xa is the original vector of attention factor, Wa and ba are the weight matrix and bias of attention layer calculation, σa is the activation function of the layer, softmax is the softmax normalization function, Ta is the attention factor vector, which can also be expressed as

[0014] Ta=a1,a2,…,an

[0015] The calculation method of the fully connected layer is

[0016] XT=a1x1,a2x2,…anxn

[0017] Xc=σcWcXT+bc

[0018] Where σc and bc are the weight matrix and bias of the attention layer, σc is the activation function of the layer, and Xc is the final predicted value of the remaining charging time.

[0019] Compared with existing technologies, this invention offers the following advantages: The robot's charging auxiliary monitoring and full-charge automatic stop decision method automatically stops the robot's charging process, preventing battery damage and cable heating caused by prolonged charging. Flexible switching between charging stages extends the robot's battery life. When predicting the remaining time in each charging stage, an attention-based backpropagation algorithm is used to optimize backpropagation, achieving good prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of the working principle of this algorithm; DETAILED DESCRIPTION

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are intended only to illustrate the present invention and are not intended to limit the scope of the present invention. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inward" and "outward" refer to directions toward or away from the geometric center of a particular component, respectively.

[0022] like Figure 1 The invention discloses an auxiliary monitoring and full-charge automatic stop decision method for robot charging, characterized in that: after the robot to be charged issues a charging request, the charging system collects and analyzes the initial battery data of the robot, provides a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging, and then starts charging. The charging process involves three steps: a pre-charging stage, in which the charging voltage and current are detected to determine whether they meet the charging requirements of the robot. If they do not meet the requirements, the robot charging battery information is re-collected, and then a charging plan and time are re-given. After the charging requirements are detected in the pre-charging stage, the robot waits for the pre-charging time to arrive and then enters the fast charging stage. After the fast charging stage time arrives, the floating charging stage begins. After the floating charging stage time arrives, the charging system terminal collects the robot battery information and then compares it with the battery calibration parameters to determine whether the battery is full. After it is full, the automatic stop decision is made to avoid damage to the robot battery due to overcharging. During the charging process, the charging terminal monitors the charging voltage, current and other parameters in real time to avoid battery damage due to abnormal voltage and current.

[0023] As an embodiment of the present invention, Figure 1As shown, the auxiliary monitoring and full-stop decision-making method for robot charging is that after the robot to be charged sends a charging request, the charging system collects and analyzes the initial battery data of the robot, gives a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging. The charging time includes the pre-charging stage time, the fast charging stage time and the floating charging stage time, and then charging begins. The charging process involves three processes, the pre-charging stage, the fast charging stage and the floating charging stage; in the pre-charging stage, the charging voltage and current are detected to determine whether they meet the robot charging requirements. If they do not meet the requirements, the robot charging battery information is collected again, and then the charging plan and time are re-given. After the pre-charging stage detects that the charging requirements are met, it is determined whether the pre-charging time has reached the pre-charging time, and the fast charging stage is entered after the pre-charging time has been reached; in the fast charging stage, the system performs timing to determine whether the estimated fast charging time has arrived. During the period of time, after reaching the fast charging stage, it enters the floating charging stage. During the floating charging stage, the system determines whether the floating charging time has arrived. After the floating charging time is reached, the charging system terminal collects the robot battery information, and then compares it with the battery calibration parameters to determine whether the battery is full. After it is full, it makes an automatic stop decision. If it is judged that it is not full, the battery information is fed back to the system. The system calculates the charging parameters and time of the robot again, and then charges it, so as to avoid damage to the robot battery due to overcharging. During the charging process, the charging terminal monitors the charging voltage, current and other parameters in real time to avoid battery damage due to abnormal voltage and current. The robot charging auxiliary monitoring and full automatic stop decision method are used to perform statistical analysis and record the charging data of different robots during the charging process, and optimize the robot charging automatic stop decision plan through autonomous learning, so that the robot uses the best charging automatic stop plan each time it is charged.

[0024] As an embodiment of the present invention, the robot charging auxiliary monitoring and full-charge automatic stop decision method has a smooth charging state switching mechanism: at the end of the pre-charge phase, a prediction model is used to predict the remaining time to reach a specified starting SOC of the rapid charge phase. If the predicted time is less than the pre-charge-to-rapid charge switching threshold, the charging current is adjusted to half the sum of the stable currents of the pre-charge and rapid charge phases. At the end of the rapid charge phase, a prediction model is used to predict the remaining time to reach a specified full-charge SOC. If the predicted time is less than the rapid charge-to-full charge switching threshold, the charging current is adjusted to half the stable current of the rapid charge phase.

[0025] As a preferred technical solution of the present invention, the charging state soft switching mechanism prediction model is a remaining charging time prediction model based on the attention mechanism to improve the BP algorithm. The specific prediction method is: first, the battery information including voltage, current, SOC, etc. in the past time window is extracted into a feature vector X:

[0026] X=x1,x2,x3,…xn

[0027] In the formula, n is the length of the vector. The attention mechanism is constructed to improve the BP algorithm, which consists of an attention mechanism layer and a fully connected layer with n input nodes and 1 output node. The calculation method of the attention mechanism layer is

[0028] Xa=σaWaX+ba

[0029] Ta=softmaxXa

[0030] Where Xa is the original vector of attention factor, Wa and ba are the weight matrix and bias of attention layer calculation, σa is the activation function of the layer, softmax is the softmax normalization function, Ta is the attention factor vector, which can also be expressed as

[0031] Ta=a1,a2,…,an

[0032] The calculation method of the fully connected layer is

[0033] XT=a1x1,a2x2,…anxn

[0034] Xc=σcWcXT+bc

[0035] Where σc and bc are the weight matrix and bias of the attention layer, σc is the activation function of the layer, and Xc is the final predicted value of the remaining charging time.

[0036] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A robot charging auxiliary monitoring and full-time automatic stop decision method, characterized by: The auxiliary monitoring and full-time automatic stop decision method for robot charging is that after the robot to be charged sends a charging request, the charging system collects and analyzes the initial battery data of the robot, gives a reasonable charging plan, and calculates and estimates the charging time required for the robot to complete charging, and then starts charging. The charging process involves three processes: a pre-charging stage, in which the charging voltage and current are detected to determine whether they meet the charging requirements of the robot. If they do not meet the requirements, the robot charging battery information is re-collected, and then a charging plan and time are re-given. After the pre-charging stage detects that the charging requirements are met, the pre-charging time is waited for and then the fast charging stage is entered. After the fast charging stage time is reached, the floating charging stage is entered. After the floating charging stage time is reached, the charging system terminal collects the robot battery information, and then compares it with the battery calibration parameters to determine whether the battery is full. After it is full, a self-stop decision is made to avoid damage to the robot battery due to overcharging. During the charging process, the charging terminal monitors the charging voltage and current parameters in real time to avoid damage to the battery due to abnormal voltage and current. The charging state soft switching mechanism of the auxiliary monitoring and full-charge automatic stop decision method for robot charging is as follows: at the end of the pre-charging stage, the prediction model is used to predict the remaining time to reach the specified starting SOC of the fast charging stage. When the predicted time is less than the pre-charging-fast charging switching threshold, the charging current is adjusted to half of the sum of the stable currents of the pre-charging and fast charging stages. At the end of the fast charging stage, the prediction model is used to predict the remaining time to reach the specified full-charge SOC. When the predicted time is less than the fast-charging-full-charge switching threshold, the charging current is adjusted to half of the stable current of the fast charging stage. The charging state soft switching mechanism prediction model is a remaining charging time prediction model based on the attention mechanism to improve the BP algorithm. The specific prediction method is: first, the battery information including voltage, current, and SOC in the past time window is extracted into a feature vector : ； Where, For vector length, an attention mechanism is constructed to improve the BP algorithm, which consists of an attention mechanism layer and a fully connected layer with n input nodes and 1 output node. The calculation method of the attention mechanism layer is ； ； Where, is the original vector of attention factor, and Calculate the weight matrix and bias for the attention layer, is the activation function of this layer, for Normalization function, is the attention factor vector, which can also be expressed as ； The calculation method of the fully connected layer is ； ； Where, and Calculate the weight matrix and bias for the attention layer, is the activation function of this layer, To finally obtain the remaining charging time prediction value.

2. The robot charging auxiliary monitoring and full-charge automatic stop decision method according to claim 1, characterized in that: The robot charging auxiliary monitoring and full-charge automatic stop decision method performs statistical analysis and records on the charging data of different robots during the charging process, and optimizes the robot charging automatic stop decision plan through autonomous learning, so that the robot uses the best charging automatic stop plan each time it charges.

3. The robot charging auxiliary monitoring and full-charge automatic stop decision method according to claim 1, characterized in that: The robot charging auxiliary monitoring and full-stop decision method collects robot battery information and estimates the time for full charging. The charging time includes pre-charging stage time, fast charging stage time and floating charging stage time.

Images