Water hydrogen production platform robot walking positioning method

Abstract

According to the water hydrogen production platform robot walking positioning method provided by the invention, high-precision walking positioning is realized through a comprehensive prediction model, error calculation and correction, application of an IMU inertial unit, inertial measurement parameter adjustment and a track plotting method. The method comprises the following specific steps: initializing a system, collecting data, estimating parameters, comparing a predicted value with a planned route, processing an error, judging whether the error exceeds the standard or not, correcting and adjusting the parameters, and judging whether a task is completed or not. State prediction and error correction are carried out on the direction angle and the moving speed through a Kalman filtering algorithm, so that the accuracy of a positioning result is improved; the system is suitable for various application scenes such as a hydrogen fuel cell system water hydrogen production platform, a solar power generation system water hydrogen production platform and a wind power generation system water hydrogen production platform, the working efficiency and accuracy of the water hydrogen production process can be effectively improved, the maintenance cost is reduced, and the working time is shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of robot walking and positioning, and in particular to a robot walking and positioning method for a water hydrogen production platform. Background Art

[0002] The water hydrogen production platform is a kind of equipment that uses the principle of water electrolysis to produce hydrogen, electrolyzes water into hydrogen and oxygen, and separates them. The water hydrogen production platform robot is an important part of the platform. It has independent perception, decision-making and execution capabilities, and can perform tasks without human intervention, improving the efficiency of hydrogen production and the intelligence and automation of the platform. The walking positioning module in the robot mainly relies on inertial measurement unit (IMU) technology, and obtains motion state parameters such as speed, direction angle and position through inertial sensors.

[0003] However, the existing IMU technology has problems with instantaneous measurement of moving speed and azimuth, making it difficult to perform long-term accurate calculations, resulting in accumulated azimuth errors and positional offsets. In addition, for large hydrogen production equipment, the equipment is large in size, complex in structure, and has a long moving trajectory, requiring multiple sensors to achieve effective motion control. Although the existing technical solution has optimized the moving mechanism, problems such as excessive azimuth errors and positional offsets may still occur in actual operation.

[0004] The main shortcoming of existing IMU technology is that the calculation of moving speed and azimuth angle mainly relies on instantaneous measurement, and it is impossible to perform accurate calculations for a long time, resulting in accumulated azimuth angle errors and position offsets. For large-scale hydrogen production equipment, the equipment is large in size, complex in structure, and has a long moving trajectory, and multiple sensors are required to achieve effective motion control. Although the existing technical solutions have optimized the moving mechanism, in actual operation, there may still be excessive azimuth angle errors and position offsets, affecting the stability and safety of the equipment. Therefore, there is an urgent need for a walking positioning technology that can accurately calculate moving parameters for a long time and improve the moving accuracy and stability of large-scale hydrogen production equipment. Summary of the invention

[0005] In view of the technical problems existing in the background technology, the present invention provides a walking and positioning method for a water hydrogen production platform robot; a walking and positioning technology for a water hydrogen production platform robot that improves the movement accuracy and stability of large-scale hydrogen production equipment and reduces angular error and position offset.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions to achieve: A walking and positioning method for a water hydrogen production platform robot comprises the following steps: S1. Collect the vehicle's forward direction angle and forward speed through the inertial measurement unit (IMU) to calculate the distance between the current position and the next position; S2, predicting the forward direction angle, forward speed and current position, and obtaining a predicted value and a predicted error; S3, comparing the predicted value with the preset planned route, calculating the position of the current vehicle in the planned route and the distance from the current position as error processing parameters; S4. When the predicted forward direction angle error and speed error are too large, the collection of the forward direction angle and speed parameters is stopped, and the early warning mechanism is activated; S5. If the difference between the current position and the actual position exceeds the set threshold, the estimation of the current position is stopped to prevent the position error from affecting the robot's walking; S6. To adjust the inertial measurement parameters of the robot, it is necessary to collect more inertial measurement data for calculation to improve the prediction accuracy; S7. Use the dead reckoning algorithm to obtain the position, direction angle and moving speed of the robot, obtain an accurate state parameter matrix, and calibrate and adjust the parameters.

[0007] In a preferred solution, in S2, the forward direction angle prediction step includes: 2.1. The inertial measurement unit (IMU) continuously collects three heading angles, the average of which is equal to the heading angle of the current position; 2.2. Assume that the two most recent direction angles are A1 and A2, and the actual direction angle measured at the current moment is A3. The direction angles predicted or calculated by the model are A1*, A2*, and A3* respectively; their corresponding standard errors are (A3-A3*)², (A2- A2*)², and (A1- A1*)²; 2.3. Taking the current position angle as the standard angle A0, and the angle of the current acquisition point as Ax, the calculation formula for the predicted angle Ax* is: Ax*=f(A0,A1,A2,(A3- A3*)²,(A2- A2*)²,(A1- A1*)²); Where: f represents the polynomial regression model, and its specific form is: Ax*=w0+w1A0+w2A1+ w3A2+w4(A3- A3*)²+w5(A2- A2*)²+w6(A1- A1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

[0008] In a preferred solution, in S2, the forward speed prediction step includes: 3.1. The inertial measurement unit (IMU) continuously collects three forward speeds, the average of which is equal to the moving speed of the current position; 3.2. Assume that the forward speeds collected in the last two times are V1 and V2 respectively, and the actual speed measurement value at the current moment is V3. The speeds predicted or calculated by the model are V1*, V2*, and V3* respectively, and their corresponding standard errors are (V3- V3*)², (V2- V2*)², and (V1- V1*)²; 3.3. Taking the moving speed of the current position as the standard moving speed V0, assuming the moving speed of the current collection point is Vx, the calculation formula for the predicted moving speed Vx* is: Vx*=f(V0,V1,V2,(V3- V3*)²,(V2- V2*)²,(V1-V1*)²); Where: f represents the polynomial regression model, and its specific form is: Vx*=w0+w1V0+w2V1+ w3V2+w4(V3-V3*)²+w5(V2- V2*)²+w6(V1- V1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

[0009] In a preferred solution, in S2, the current position prediction step includes: 4.1. The inertial measurement unit (IMU) continuously collects the distance between three collection points, and the average value is equal to the distance of the current position; 4.2. Let the distances between the two most recent acquisition points be X1 and X2, let the actual measured value of the current position be X3, and the positions predicted or calculated by the model be X1*, X2*, and X3*, respectively. The corresponding standard errors are (X3- X3*)², (X2- X2*)², and (X1- X1*)²; 4.3. Given the known distance X0 between the first two acquisition points, let the predicted position value Xx at the current moment be calculated as: Xx*=f(X0,X1,X2,(X3- X3*)²,(X2- X2*)²,(X1-X1*)²); Where: f represents the polynomial regression model, and its specific form is: Xx*=w0+w1X0+w2X1+ w3X2+w4(X3- X3*)²+w5(X2- X2*)²+w6(X1- X1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

[0010] In a preferred solution, in S6, the robot inertial measurement parameter adjustment step includes: 5.1. Use three inertial measurement units (IMU1, IMU2, IMU3) to collect data simultaneously; 5.2. Substitute the collected data into the inertial measurement unit model to calculate the changes in direction angle, moving speed and position; the inertial measurement unit model is as follows: X=Aω+Ba+C; Where: X is the displacement matrix, ω is the angular velocity output matrix, a is the acceleration output matrix, A and B are the coefficient matrices of the angular velocity gyroscope and accelerometer respectively, and C is a constant term; 5.3. Calculate the standard error of direction angle, moving speed and position. The standard error formula is: F1=(A3- A3*)²; F2=(V1-V1*)²; F3=(X3- X3*)²; in, F 1. F 2. F 3 are the standard errors of direction angle, moving speed and position respectively.

[0011] In a preferred solution, in S7, the robot parameter correction and adjustment step includes: 6.1. Use dead reckoning to calculate the robot’s position, direction angle, and moving speed; 6.2. Obtain accurate state parameter matrix, and calibrate and adjust the parameters.

[0012] In a preferred solution, when the robot is walking, the forward direction angles of the current moving position and the next position are determined to ensure that the robot does not deviate from the planned path; if the forward direction angles are inconsistent, the robot temporarily stops walking and sends an abnormal signal.

[0013] A walking positioning method for a water hydrogen production platform robot, the beneficial effects of the method include but are not limited to the following points: 1. The robot positioning results are corrected by using machine learning technology, especially the Kalman filter algorithm is used to effectively correct the errors measured by the inertial sensor, which significantly improves the robot's positioning accuracy. This method not only corrects the measurement deviation caused by the limitations of the sensor itself, but also optimizes the robot's self-positioning ability, ensuring that the robot can reach the target position more accurately when performing tasks; 2. The present invention not only improves the stability of the robot, but also enhances the overall reliability of the system. Through real-time state prediction and error analysis, it can timely discover and correct possible deviations, reduce the uncertainty and risks in the positioning process, and ensure the safe and efficient operation of the water hydrogen production process; 3. The present invention achieves high-precision measurement of the direction angle and moving speed by predicting the state of these two important parameters and calculating the corresponding prediction errors; based on these precise measurement results, the robot's action strategy can be further adjusted to make its direction angle and moving speed more in line with expectations, thereby improving the accuracy of subsequent position predictions; 4. Due to the use of predictive models and error correction technology, the robot can move on the platform more accurately and reliably according to the calculated position state parameters, reducing the need for maintenance due to positioning errors and lowering long-term operating costs. In addition, the robot's efficient actions directly promote the efficiency of the water-to-hydrogen process, shorten the work cycle, and improve the overall production capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic diagram of the overall operation mode of the present invention. DETAILED DESCRIPTION

[0015] The present invention provides a walking positioning method for a water hydrogen production platform robot, which realizes high-precision walking positioning through a comprehensive prediction model, error calculation and correction, application of an IMU inertial unit, adjustment of inertial measurement parameters and dead reckoning. The following are specific embodiments, showing the application of the present invention in different scenarios.

[0016] Embodiment 1: A water hydrogen production platform for a hydrogen fuel cell system with an error value that does not exceed the standard; 1. Start the robot, initialize the IMU unit and other sensors, set the initial position and orientation angle, and ensure that all sensors work properly. The initial position is (x0, y0) = (0, 0) meters, and the initial orientation angle is A0 = 0°. 2. The IMU unit continuously collects the robot's forward direction angle and forward speed, and calculates the distance between the current position and the next position; The collected data include direction angle A0=0°, A1=1°, A2=2°; moving speed V0=0.5m / s, V1=0.55m / s, V2=0.6m / s; and position X0=0m, X1=0.5m, X2=1.0m; 3. Use the direction angle prediction formula Ax*=f(A0,A1,A2,(A3- A3*)²,(A2- A2*)²,(A1- A1*)²); calculate the current direction angle prediction value Ax*; assuming A3=3°, then Ax*=3.1°; Use the moving speed prediction formula Vx*=f(V0,V1,V2,(V3- V3*)²,(V2- V2*)²,(V1-V1*)²); calculate the current moving speed prediction value Vx*; assuming Vx=0.65m / s, then Vx*=0.66m / s; Use the position prediction formula Xx*=f(X0,X1,X2,(X3- X3*)²,(X2- X2*)²,(X1-X1*)²); Calculate the current predicted moving speed value Xx*; assuming Xx=1.5 meters, then Xx*=1.51 meters; 4. Compare the predicted value with the preset planned route, calculate the position of the current vehicle in the planned route and the distance from the current position as error processing parameters; 5. Calculate the error between the predicted value and the actual value of the direction angle F1 = (A3- A3*)² = (3-3.1) 2 =0.01; Calculate the error between the predicted value and the actual value of the moving speed F2 = (V3- V3*)² = (0.65-0.66) 2 =0.0001; Calculate the error between the predicted position and the actual value F3 = (X3- X3*)² = (1.5-1.51) 2 =0.0001; 6. Since F1=0.01, F2=0.0001, and F3=0.0001 are all less than the set threshold, proceed to the next step; 7. Determine whether the task is completed. If the task is completed, end the process. If the task is not completed, return to step 2 to continue collecting data.

[0017] Embodiment 2: Wind power generation system water hydrogen production platform with error value exceeding the standard; 1. Start the robot, initialize the IMU unit and other sensors, set the initial position and orientation angle, and ensure that all sensors work properly. The initial position is (x0, y0) = (0, 0) meters, and the initial orientation angle is A0 = 0°. 2. The IMU unit continuously collects the robot's forward direction angle and forward speed, and calculates the distance between the current position and the next position; The collected data include direction angle A0=0°, A1=5°, A2=10°; moving speed V0=0.3m / s, V1=0.35m / s, V2=0.4m / s; and position X0=0m, X1=0.3m, X2=0.6m; 3. Use the direction angle prediction formula Ax*=f(A0,A1,A2,(A3- A3*)²,(A2- A2*)²,(A1- A1*)²); calculate the current direction angle prediction value Ax*; assuming A3=15°, then Ax*=20°; Use the moving speed prediction formula Vx*=f(V0,V1,V2,(V3- V3*)²,(V2- V2*)²,(V1-V1*)²); calculate the current moving speed prediction value Vx*; assuming Vx=0.5m / s, then Vx*=0.7m / s; Use the position prediction formula Xx*=f(X0,X1,X2,(X3- X3*)²,(X2- X2*)²,(X1-X1*)²); Calculate the current moving speed prediction value Xx*; assuming Xx=0.9m, then Xx*=1.0m; 4. Compare the predicted value with the preset planned route, calculate the position of the current vehicle in the planned route and the distance from the current position as error processing parameters; 5. Calculate the error between the predicted value and the actual value of the direction angle F1 = (A3- A3*)² = (15-20) 2 =25; Calculate the error between the predicted value and the actual value of the moving speed F2 = (V3- V3*)² = (0.5-0.7) 2 =0.04; Calculate the error between the predicted position and the actual value F3 = (X3- X3*)² = (0.9-1.0) 2 =0.01; 6. Since F1=25 is greater than the set threshold of 10°, F2=0.04 is greater than the set threshold of 0.01m / s, and F3=0.01 does not exceed the set threshold of 0.1m, the warning mechanism is triggered and the collection of positioning parameters is stopped; 7. Since the angular error F1=25 and the moving speed error F2=0.04 both exceed the set thresholds, the early warning mechanism is activated, the collection of positioning parameters is stopped, the sensor and system status are checked, and the positioning process is restarted after the fault is eliminated.

[0018] It is obvious to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be regarded as exemplary and non-restrictive from all points of view, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention; any reference numerals in the claims should not be regarded as limiting the claims involved.

Claims

1. A walking and positioning method for a water hydrogen production platform robot, characterized in that: The following steps are involved: S1. Collect the vehicle's forward direction angle and forward speed through the inertial measurement unit (IMU) to calculate the distance between the current position and the next position; S2, predicting the forward direction angle, forward speed and current position, and obtaining a predicted value and a predicted error; S3, comparing the predicted value with the preset planned route, calculating the position of the current vehicle in the planned route and the distance from the current position as error processing parameters; S4. When the predicted forward direction angle error and speed error are too large, the collection of the forward direction angle and speed parameters is stopped, and the early warning mechanism is activated; S5. If the difference between the current position and the actual position exceeds the set threshold, the estimation of the current position is stopped to prevent the position error from affecting the robot's walking; S6. To adjust the inertial measurement parameters of the robot, it is necessary to collect more inertial measurement data for calculation to improve the prediction accuracy; S7. Use the dead reckoning algorithm to obtain the position, direction angle and moving speed of the robot, obtain an accurate state parameter matrix, and calibrate and adjust the parameters.

2. The walking and positioning method of the water hydrogen production platform robot according to claim 1 is characterized in that: In S2, the heading angle prediction step includes: 2.

1. The inertial measurement unit (IMU) continuously collects three heading angles, the average of which is equal to the heading angle of the current position; 2.

2. Assume that the two most recent direction angles are A1 and A2, and the actual direction angle measured at the current moment is A3. The direction angles predicted or calculated by the model are A1*, A2*, and A3* respectively; their corresponding standard errors are (A3- A3*)², (A2- A2*)², and (A1- A1*)²; 2.

3. Taking the current position angle as the standard angle A0, and the angle of the current acquisition point as Ax, the calculation formula for the predicted angle Ax* is: Ax*=f(A0,A1,A2,(A3- A3*)²,(A2- A2*)²,(A1- A1*)²); Where: f represents the polynomial regression model, and its specific form is: Ax*=w0+w1A0+w2A1+ w3A2+w4(A3- A3*)²+w5(A2- A2*)²+w6(A1- A1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

3. The walking and positioning method of the water hydrogen production platform robot according to claim 1, characterized in that: In S2, the forward speed prediction step includes: 3.

1. The inertial measurement unit (IMU) continuously collects three forward speeds, the average of which is equal to the moving speed of the current position; 3.

2. Assume that the forward speeds of the two most recent collections are V1 and V2, and the actual speed measurement value at the current moment is V3. The speeds predicted or calculated by the model are V1*, V2*, and V3*, and their corresponding standard errors are (V3- V3*)², (V2-V2*)², and (V1- V1*)²; 3.

3. Taking the moving speed of the current position as the standard moving speed V0, assuming the moving speed of the current collection point is Vx, the calculation formula for the predicted moving speed Vx* is: Vx*=f(V0,V1,V2,(V3- V3*)²,(V2- V2*)²,(V1-V1*)²); Where: f represents the polynomial regression model, and its specific form is: Vx*=w0+w1V0+w2V1+ w3V2+w4(V3-V3*)²+w5(V2- V2*)²+w6(V1- V1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

4. The walking and positioning method of the water hydrogen production platform robot according to claim 1, characterized in that: In S2, the current position prediction step includes: 4.

1. The inertial measurement unit (IMU) continuously collects the distance between three collection points, and the average value is equal to the distance of the current position; 4.

2. Let the distances between the two most recent acquisition points be X1 and X2, let the actual measured value of the current position be X3, and the positions predicted or calculated by the model be X1*, X2*, and X3*, respectively. The corresponding standard errors are (X3- X3*)², (X2-X2*)², and (X1- X1*)²; 4.

3. Given the known distance X0 between the first two acquisition points, let the predicted position value Xx at the current moment be calculated as: Xx*=f(X0,X1,X2,(X3- X3*)²,(X2- X2*)²,(X1-X1*)²); Where: f represents the polynomial regression model, and its specific form is: Xx*=w0+w1X0+w2X1+ w3X2+w4(X3- X3*)²+w5(X2- X2*)²+w6(X1- X1*)²; Where: w0, w1, w2, w3, w4, w5 and w6 are the weight parameters of the model, which can be trained and determined by minimizing the error between the predicted value and the actual value.

5. The walking and positioning method of the water hydrogen production platform robot according to claim 1, characterized in that: In S6, the robot inertial measurement parameter adjustment step includes: 5.

1. Use three inertial measurement units (IMU1, IMU2, IMU3) to collect data simultaneously; 5.

2. Substitute the collected data into the inertial measurement unit model to calculate the changes in direction angle, moving speed and position; the inertial measurement unit model is as follows: X=Aω+Ba+C; Where: X is the displacement matrix, ω is the angular velocity output matrix, a is the acceleration output matrix, A and B are the coefficient matrices of the angular velocity gyroscope and accelerometer respectively, and C is a constant term; 5.

3. Calculate the standard error of direction angle, moving speed and position. The standard error formula is: F1=(A3- A3*)²; F2=(V1-V1*)²; F3=(X3- X3*)²; in, F 1. F 2. F 3 are the standard errors of direction angle, moving speed and position respectively.

6. The walking and positioning method of the water hydrogen production platform robot according to claim 1, characterized in that: In S7, the robot parameter correction and adjustment steps include: 6.

1. Use dead reckoning to calculate the robot’s position, direction angle and moving speed; 6.

2. Obtain accurate state parameter matrix, and calibrate and adjust the parameters.

7. The walking and positioning method of a water hydrogen production platform robot according to any one of claims 1 to 6, characterized in that: When the robot is walking, it determines the forward direction angle between the current position and the next position to ensure that the robot does not deviate from the planned path; if the forward direction angle is inconsistent, the robot temporarily stops walking and sends an abnormal signal.

Images