A robot walking module voltage feedback gait adjusting method and system

Abstract

The application relates to a robot walking module voltage feedback gait adjustment method and system, and relates to the technical field of robot gait control, which comprises the following steps: collecting real-time discharge current and real-time battery terminal voltage of a preset robot walking module; analyzing the real-time discharge current and the real-time battery terminal voltage to generate a voltage drop trend prediction curve; analyzing the voltage drop trend prediction curve and preset reference robot gait parameters to generate actual robot gait parameters; controlling the robot walking module to walk according to the actual robot gait parameters, collecting real-time joint position signals and real-time foot end contact forces; analyzing the real-time joint position signals and the real-time foot end contact forces to generate joint current amplitude limiting; supplying power to the robot walking module according to the joint current amplitude limiting, and continuing to control the robot walking module to walk according to the actual robot gait parameters. The application has the effect of improving gait adjustment stability.

Description

Technical Field

[0001] This application relates to the technical field of robot gait control, and in particular to a voltage feedback gait adjustment method and system for a robot walking module. Background Technology

[0002] A robot walking module refers to the collection of hardware in a robot that enables the robot to move. It mainly consists of a battery, leg mechanical structure, joint drive motors, and leg sensors.

[0003] In related technologies, gait regulation methods based on voltage feedback typically employ a passive response mechanism: real-time monitoring of the bus voltage, and when the voltage drops to a preset threshold, uniformly reducing the power output of all joint motors, or triggering a protective frequency and amplitude reduction strategy to slow down battery consumption. Some solutions further introduce battery state of charge estimation, adjusting gait parameters in segments based on the current remaining charge.

[0004] Regarding the aforementioned technologies, on the one hand, the control strategy relies on real-time passive feedback of the voltage signal, and protective regulation is only activated when the bus voltage actually drops. This lag response can easily cause gait abrupt changes, jamming, or even falls when the torque demand of the support leg increases instantaneously. On the other hand, conventional power reduction protection measures usually apply uniform voltage reduction to all joint motors without distinguishing between the distinct torque demands of the support phase and the swing phase in the gait cycle. This results in the support leg becoming unstable due to the average reduction of power, while the swing leg still has power redundancy, leading to an unreasonable distribution of energy efficiency in the entire machine, and there is still room for improvement. Summary of the Invention

[0005] To improve the stability of gait adjustment, this application provides a voltage feedback gait adjustment method and system for a robot walking module.

[0006] In a first aspect, this application provides a voltage feedback gait adjustment method for a robot walking module, which adopts the following technical solution: A method for voltage feedback gait adjustment in a robot walking module includes: Collect the real-time discharge current and real-time battery terminal voltage of the preset robot walking module; The real-time discharge current and real-time battery terminal voltage are analyzed to generate a voltage drop trend prediction curve. The voltage drop trend prediction curve and the preset baseline robot gait parameters are analyzed to generate the actual robot gait parameters. The robot's walking module is controlled to walk based on the actual robot gait parameters, and real-time joint position signals and real-time foot contact force are collected. The real-time joint position signal and real-time foot contact force are analyzed to generate joint current limiting. Power is supplied to the robot's walking module based on the joint current limit, and the robot's walking module continues to move according to the actual robot gait parameters.

[0007] Optionally, the step of analyzing the real-time discharge current and real-time battery terminal voltage to generate a voltage drop trend prediction curve includes: The real-time discharge current is input into the preset state equation for calculation to generate the uncorrected remaining charge curve and polarization voltage change curve. The uncorrected remaining charge curve and polarization voltage change curve are input into the preset observation equation for calculation to generate the theoretical battery terminal voltage. Calculate the difference between the real-time battery terminal voltage and the theoretical battery terminal voltage to generate the battery terminal voltage residual; The uncorrected remaining power curve is corrected based on the preset Kalman gain coefficient and battery terminal voltage residual to generate the optimal remaining power change curve. Based on the preset sampling length, the remaining power time curve and the remaining power are extracted from the optimal remaining power change curve. Calculate the average of the sampled remaining battery power to generate the average remaining battery power; The sampling remaining power time curve and average remaining power are analyzed to generate a voltage drop trend prediction curve.

[0008] Optionally, the step of analyzing the sampled remaining power time curve and the average remaining power to generate a voltage drop trend prediction curve includes: The sampled remaining power time curve and average remaining power are substituted into the preset remaining power decay rate calculation formula to generate the average remaining power decay rate. Extract the latest remaining battery capacity from the optimal remaining capacity change curve; The product of the average remaining battery capacity decay rate and the preset prediction period is summed with the latest remaining battery capacity to generate the predicted remaining battery capacity. The sampling length and real-time discharge current are analyzed to generate an average discharge current; The predicted remaining battery capacity and average discharge current are substituted into a preset battery terminal voltage prediction formula to generate a voltage drop trend prediction curve.

[0009] Optionally, the step of analyzing the voltage drop trend prediction curve and preset benchmark robot gait parameters to generate actual robot gait parameters includes: Extract the lowest predicted voltage from the voltage drop trend prediction curve; Calculate the difference between the preset battery high-level threshold and the lowest predicted voltage to generate the battery's maximum drop voltage; Calculate the quotient of the battery's maximum voltage drop and the preset difference between the normal battery's high and low voltages to generate the cycle gait adjustment coefficient. Collect the gait adjustment coefficient from the previous cycle; Calculate the difference between the gait adjustment coefficient of the current cycle and the gait adjustment coefficient of the previous cycle to generate the gait adjustment coefficient difference between cycles; The difference in gait adjustment coefficients during the cycle and the gait adjustment coefficients of the previous cycle are substituted into the preset formula for calculating smooth gait adjustment parameters to generate the actual gait adjustment coefficients. The baseline robot gait parameters and actual gait adjustment coefficients are analyzed to generate the actual robot gait parameters.

[0010] Optionally, the steps of analyzing the baseline robot gait parameters and the actual gait adjustment coefficients to generate the actual robot gait parameters include: Find the maximum step frequency, the reference step frequency difference, the maximum step length, the reference step length difference, the maximum support phase duration percentage, and the reference support phase duration percentage difference in the reference robot gait parameters; Calculate the product of the actual gait adjustment coefficient with the reference step frequency difference, reference step length difference, and reference support phase duration percentage difference to generate the step frequency adjustment amount, step length adjustment amount, and support phase duration percentage adjustment amount; Calculate the difference between the maximum step frequency and the step frequency adjustment amount to generate the actual robot step frequency; Calculate the difference between the maximum step size and the step size adjustment to generate the actual robot step size; Calculate the difference between the maximum support phase duration percentage and the adjustment amount of the support phase duration percentage to generate the actual robot support phase duration percentage; The actual robot's step frequency, actual robot step length, and actual robot support phase duration are summarized to generate actual robot gait parameters.

[0011] Optionally, the step of analyzing real-time joint position signals and real-time foot contact forces to generate joint current limiting includes: The real-time joint position signal is calculated based on a preset kinematic model to generate the forward direction component of the foot. The forward direction component of the foot, the real-time contact force of the foot, and the preset support swing side characteristic parameters are analyzed to generate the motion types of each side of the foot. The power allocation weights for each joint are determined from the preset power allocation weights based on the movement type of each side of the foot. Collect the maximum safe discharge current; Calculate the product of the real-time battery terminal voltage and the maximum safe discharge current to generate the maximum usable power; The power allocation weights and maximum available power of each joint are analyzed to generate joint current limits.

[0012] Optionally, the step of analyzing the power allocation weights and maximum available power of each joint to generate joint current limits includes: Calculate the sum of the power allocation weights for each joint to generate the total power allocation weights; Calculate the quotient of the power allocation weight of each joint to the total power allocation weight to generate the power allocation ratio of each joint; Calculate the product of the power allocation ratio of each joint and the maximum available power to generate the power allocation for each joint; Calculate the quotient of the power allocated to each joint and the real-time battery terminal voltage to generate the joint current limit.

[0013] Secondly, this application provides a voltage feedback gait adjustment system for a robot walking module, which adopts the following technical solution: A voltage feedback gait adjustment system for a robot walking module includes: The acquisition module is used to acquire real-time discharge current, real-time battery terminal voltage, real-time joint position signal, and real-time foot contact force. A memory for storing a program for a voltage feedback gait adjustment method for a robot walking module as described in any of the preceding claims; The processor and the program in the memory can be loaded and executed by the processor to implement a voltage feedback gait adjustment method for a robot walking module as described in any of the above.

[0014] In summary, this application includes at least one of the following beneficial technical effects: 1. By analyzing the real-time discharge current and real-time battery terminal voltage, a voltage drop trend prediction curve is obtained. The actual robot gait parameters are obtained by analyzing the voltage drop trend prediction curve and the reference robot gait parameters. The robot walking module is controlled to walk according to the actual robot gait parameters. Real-time joint position signals and real-time foot contact forces are collected. The joint current limit is obtained by analyzing the real-time joint position signals and real-time foot contact forces. The robot walking module is powered according to the joint current limit. The robot walking module is controlled to walk according to the actual robot gait parameters, thereby improving the stability of the robot walking module during gait adjustment. 2. By extracting the lowest predicted voltage from the voltage drop trend prediction curve, the difference between the battery high-level threshold and the lowest predicted voltage is calculated to obtain the maximum voltage drop of the battery. The quotient of the maximum voltage drop of the battery and the difference between the normal high and low voltages of the battery is calculated to obtain the gait adjustment coefficient for the current cycle. The difference between the gait adjustment coefficient for the current cycle and the gait adjustment coefficient for the previous cycle is calculated to obtain the gait adjustment coefficient difference during the cycle. The gait adjustment coefficient difference during the cycle and the gait adjustment coefficient for the previous cycle are substituted into the smoothing gait adjustment parameter calculation formula to calculate the actual gait adjustment coefficient. Thus, the actual robot gait parameters are obtained by analyzing the baseline robot gait parameters and the actual gait adjustment coefficient, thereby realizing proactive intervention before voltage drop and avoiding the consequences of gait abrupt changes, stuttering, or even falls as much as possible. 3. By calculating the forward direction component of the foot from the real-time joint position signal based on the kinematic model, the forward direction component of the foot, the real-time foot contact force, and the characteristic parameters of the support and swing sides are analyzed to obtain the motion type of each side of the foot. Based on the motion type of each side of the foot, the power allocation weight of each joint is found in the power allocation weight. The product of the real-time battery terminal voltage and the maximum safe discharge current is calculated to obtain the maximum available power. Thus, by analyzing the power allocation weight of each joint and the maximum available power, the joint current limit is obtained, thereby realizing the ability to distinguish the torque requirements of the support phase and the swing phase in the gait cycle, so as to rationally allocate the overall energy efficiency. Attached Figure Description

[0015] Figure 1 This is a flowchart of a voltage feedback gait adjustment method for a robot walking module according to an embodiment of this application.

[0016] Figure 2 This is a flowchart of the steps in this application embodiment to analyze the real-time discharge current and real-time battery terminal voltage to generate a voltage drop trend prediction curve.

[0017] Figure 3 This is a flowchart of the steps in this application embodiment to analyze the sampling remaining power time curve and the average remaining power to generate a voltage drop trend prediction curve.

[0018] Figure 4 This is a flowchart illustrating the steps in this application embodiment to analyze the voltage drop trend prediction curve and preset benchmark robot gait parameters to generate actual robot gait parameters.

[0019] Figure 5 This is a flowchart illustrating the steps in this application embodiment to analyze the baseline robot gait parameters and the actual gait adjustment coefficients to generate the actual robot gait parameters.

[0020] Figure 6This is a flowchart of the steps in this application embodiment to analyze real-time joint position signals and real-time foot contact force to generate joint current limiting.

[0021] Figure 7 This is a flowchart of the steps in this application embodiment to analyze the power allocation weights and maximum available power of each joint in order to generate joint current limiting. Detailed Implementation

[0022] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 7 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0023] This application discloses a voltage feedback gait adjustment method for a robot walking module. This method primarily addresses the issue of improving gait adjustment stability. Specifically, it discloses a robot walking module and a processing terminal. The processing terminal is communicatively connected to the robot walking module to achieve data interaction and control. The processing terminal analyzes the collected real-time discharge current and real-time battery terminal voltage to obtain a voltage drop trend prediction curve. It then analyzes the voltage drop trend prediction curve and benchmark robot gait parameters to obtain actual robot gait parameters. Based on these actual robot gait parameters, the processing terminal controls the robot walking module to walk. Furthermore, it analyzes real-time joint position signals and real-time foot contact force to obtain joint current limiting. Based on this joint current limiting, the processing terminal supplies power to the robot walking module and continues to control the robot walking module to walk according to the actual robot gait parameters, thereby improving the stability of the robot walking module during gait adjustment.

[0024] Reference Figure 1 This application discloses a voltage feedback gait adjustment method for a robot walking module, comprising the following steps: Step S100: Collect the preset real-time discharge current and real-time battery terminal voltage of the robot walking module.

[0025] Among them, the real-time discharge current refers to the curve of the total current value output by the battery to each joint drive motor changing over time. The real-time discharge current can reflect the instantaneous power consumption level of the robot walking module. By connecting a high-precision shunt resistor in series at the output of the battery pack, measuring the voltage drop across the shunt resistor through an isolated current detection amplifier, and finally converting it into a digital quantity through an analog-to-digital converter (ADC), the real-time discharge current can be obtained by plotting the collected current as the vertical axis and time as the horizontal axis.

[0026] Real-time battery terminal voltage refers to the curve of the total voltage value output by the battery to each joint drive motor changing over time. Real-time battery terminal voltage can reflect the current load output voltage of the battery. The battery terminal voltage is linearly reduced to a voltage that the analog-to-digital converter can withstand through a voltage divider resistor network, high-frequency noise of the motor is filtered through a low-pass filter, impedance matching is performed through a voltage follower, and finally the filtered voltage is collected by the processing terminal. The real-time battery terminal voltage can be obtained by plotting the curve with voltage as the vertical axis and time as the horizontal axis.

[0027] A robot walking module refers to the hardware assembly in a robot that enables the robot to move. It mainly consists of a battery, leg mechanical structure, joint drive motors, leg sensors for collecting relevant parameters during movement, and a communication module. After receiving relevant motion parameters from the processing terminal, the battery supplies power to the robot walking module and collects real-time data through the leg sensors. Based on the collected real-time data, the relevant motion parameters are determined, and the joint drive motors are controlled to drive the leg mechanical mechanism to move, thereby enabling the robot to walk.

[0028] Step S101: Analyze the real-time discharge current and real-time battery terminal voltage to generate a voltage drop trend prediction curve.

[0029] The voltage drop trend prediction curve refers to the future trajectory of the total output voltage calculated based on the battery's energy decay trend. This curve allows for the early detection of voltage drop magnitude and speed, enabling predictive gait adjustment of the robot's walking module. The processing terminal analyzes the real-time discharge current and real-time battery terminal voltage to obtain the voltage drop trend prediction curve. For specific methods, refer to [reference needed]. Figure 2 The steps.

[0030] Step S102: Analyze the voltage drop trend prediction curve and the preset benchmark robot gait parameters to generate actual robot gait parameters.

[0031] The actual robot gait parameters refer to the set of parameters used to drive the robot's walking module. These parameters help avoid situations where a voltage drop can cause insufficient instantaneous torque in the supporting legs of the walking module, leading to stuttering or falls during walking. The actual robot gait parameters are obtained by analyzing the voltage drop trend prediction curve and the baseline robot gait parameters at the processing terminal. Specific methods are described in [reference needed]. Figure 4 The steps.

[0032] The baseline robot gait parameters refer to the set of parameters that are pre-set to ensure that the robot maintains its basic walking ability and are used to drive the movement of the robot walking module. In one embodiment, the relevant control parameters are obtained by the operator in advance by inputting them into the processing terminal.

[0033] Step S103: Control the robot walking module to walk according to the actual robot gait parameters, and collect real-time joint position signals and real-time foot contact force.

[0034] The robot walks by controlling the joint drive motors to drive the leg mechanical mechanism based on the actual robot gait parameters. During the robot walking module's walking process, real-time joint position signals and real-time foot contact force are collected to distinguish whether each leg of the leg mechanical structure is a supporting leg or a swinging leg. Then, the joint drive motors are selectively powered according to the function of different legs.

[0035] Real-time joint position signals refer to the measured values ​​of joint rotation angles fed back by the drive motors of each joint during walking. Real-time joint position signals can accurately determine the position coordinates of the foot end of each leg mechanical structure relative to the robot's torso coordinate system. Data is collected by absolute magnetic encoders or rotary transformers in the leg sensors and then sent to the processing terminal via a serial bus to obtain real-time joint position signals.

[0036] Real-time foot contact force refers to the contact force between the foot and the ground in the leg mechanical structure of a robot walking module. Real-time foot contact force can serve as the main criterion for gait phase recognition, distinguishing whether each leg of the leg mechanical structure is a supporting leg or a swinging leg. The real-time foot contact force can be obtained by collecting data from one-dimensional or three-dimensional force sensors installed on the sole of the foot in the leg sensor.

[0037] Step S104: Analyze the real-time joint position signal and the real-time foot contact force to generate joint current limiting.

[0038] The joint current limit refers to the maximum current amplitude supplied to each joint drive motor. The joint current limit can be obtained by analyzing the real-time joint position signal and the real-time foot contact force. For specific methods, please refer to [reference needed]. Figure 6 The steps.

[0039] Step S105: Power the robot walking module according to the joint current limit, and continue to control the robot walking module to walk according to the actual robot gait parameters.

[0040] The processing terminal supplies power to the joint drive motors of the robot walking module according to the joint current limit, controls the robot walking module to walk according to the actual robot gait parameters, and continues to collect real-time discharge current and real-time battery terminal voltage to determine the actual robot gait parameters and joint current limit for the next observation window. Then, it supplies power to the robot walking module according to the joint current limit for the next observation window and controls the robot walking module to walk according to the actual robot gait parameters for the next observation window.

[0041] Reference Figure 2 The steps for analyzing real-time discharge current and real-time battery terminal voltage to generate a voltage drop trend prediction curve include: Step S200: Input the real-time discharge current into the preset state equation for calculation to generate the uncorrected remaining charge curve and polarization voltage change curve.

[0042] Among them, the uncorrected remaining charge curve refers to the curve of electrical energy changing over time obtained solely through the equation of state without incorporating real-time battery terminal voltage for feedback correction; the polarization voltage change curve refers to the curve of transient voltage component changing over time during battery charging and discharging, which is caused by the slow change of internal potential difference due to the lag of electrochemical reaction and ion diffusion rate compared to charge transfer, exhibiting characteristics similar to the charging and discharging of an RC circuit. It can reflect the internal electrochemical polarization of the battery. The processing terminal inputs the real-time discharge current into the equation of state for calculation, thereby obtaining the uncorrected remaining charge curve and the polarization voltage change curve.

[0043] The state equation is: In the formula, This refers to the uncorrected remaining battery power curve. This refers to the polarization voltage variation curve. This refers to the uncorrected remaining battery power curve from the previous moment. This refers to the real-time discharge current. This refers to the polarization voltage change curve at the previous moment. It refers to a specific sampling time within the observation window. It refers to from Time's up The duration between This refers to the coulombic efficiency of the battery. This refers to the battery's nominal capacity, which is obtained by searching the robot's basic equipment parameter database through the processing terminal. and Both refer to process noise. This refers to the process noise in the recursive state of charge calculation. This refers to the noise in the polarization voltage recursion process. This refers to the polarization resistance. This refers to polarized capacitors.

[0044] Process noise refers to the error used in the quantization recursion process. and Substitute into the formula The process noise variance of the charge state recursion is obtained, and then... The process noise variance of the polarization voltage recursion is calculated, where, This refers to the process noise variance in the recursive state of charge calculation. This refers to the minimum resolution of the current sensor. This refers to the noise variance during the polarization voltage recursion process. This refers to the peak voltage under the maximum pulse current in the polarization voltage change curve. Then, a random number generator is used to generate two random numbers with a mean of 0 and variances of the process noise variances of the state of charge recursion and polarization voltage recursion, respectively. These random numbers are the process noise of the state of charge recursion and the process noise of the polarization voltage recursion.

[0045] Polarization resistance is a physical quantity that reflects the resistance to electrode reactions and ion diffusion. It determines the magnitude of the additional polarization overpotential generated during continuous discharge. By applying a constant current pulse to the battery and then disconnecting the power, and recording the voltage rebound curve, after removing the instantaneous voltage jump caused by the ohmic internal resistance, the amplitude of the remaining slowly rebounding voltage is extracted. According to Ohm's law, the amplitude is divided by the pulse current to obtain the polarization resistance.

[0046] Polarization capacitance refers to the capacitive energy storage characteristics used to reflect the electric double layer effect and concentration polarization. It can determine whether the voltage change is instantaneous or a slow transition. Polarization resistance and polarization capacitance can determine the speed of voltage rebound or decay after a sudden change in current. After determining the polarization resistance, the least squares fit is performed on the exponential rebound curve of the voltage over time after power failure to find the optimal time constant that minimizes the error between the model and the actual measurement. Then, the polarization capacitance is obtained by dividing the time constant by the polarization resistance.

[0047] The equipment basic parameter library refers to a database used to store the rated equipment parameters of each component in the robot walking module and the relevant initial parameters preset by the operator.

[0048] Step S201: Input the uncorrected remaining charge curve and polarization voltage change curve into the preset observation equation for calculation to generate the theoretical battery terminal voltage.

[0049] The theoretical battery terminal voltage refers to the predicted terminal voltage value corresponding to the real-time discharge current at the current moment. The processing terminal can obtain the theoretical battery terminal voltage by inputting the uncorrected remaining charge curve into the observation equation.

[0050] The observation equation is: In the formula, This refers to the theoretical battery terminal voltage. This refers to the open-circuit voltage-energy function relationship. This refers to the uncorrected remaining battery power curve. This refers to the polarization voltage variation curve. This refers to the real-time discharge current. This refers to the battery's internal resistance, which the processing terminal finds in the device's basic parameter database. Taking 1kHz as an example, This refers to observed noise, in order to =0.01V 2 For example.

[0051] Step S202: Calculate the difference between the real-time battery terminal voltage and the theoretical battery terminal voltage to generate the battery terminal voltage residual.

[0052] Among them, the battery terminal voltage residual refers to the physical quantity used to reflect the deviation between the predicted voltage and the measured voltage. The processing terminal subtracts the theoretical battery terminal voltage from the real-time battery terminal voltage to obtain the battery terminal voltage residual.

[0053] Step S203: Correct the uncorrected remaining power curve based on the preset Kalman gain coefficient and battery terminal voltage residual to generate the optimal remaining power change curve.

[0054] The optimal remaining charge change curve refers to the curve of electrical energy changing over time obtained by using the state equation and incorporating real-time battery terminal voltage for feedback correction. The processing terminal organizes the uncorrected remaining charge curve and the polarization voltage change curve into a state vector. Then, by substituting the Kalman gain coefficient and the state vector into the state correction equation to correct the uncorrected remaining charge curve, the optimal remaining charge change curve can be obtained.

[0055] The state correction equation is: In the formula, This refers to the corrected state vector. This refers to the state vector. This refers to the residual voltage at the battery terminals. This refers to the Kalman gain coefficient.

[0056] The Kalman gain coefficient refers to the weight assigned to the real-time battery terminal voltage and the theoretical battery terminal voltage based on the uncertainty of the predicted voltage and the actual voltage, thereby fusing the real-time battery terminal voltage and the theoretical battery terminal voltage. First, the posterior error covariance of the previous moment is recursively derived into the prior error covariance of the current moment through process noise. Then, the prior error covariance of the current moment is mapped to the measurement space through the observation matrix to obtain the uncertainty of the model's predicted terminal voltage. This uncertainty is then added to the sensor measurement noise covariance to form the total uncertainty. Finally, the prior error covariance of the current moment, the observation matrix, and the reciprocal of the total uncertainty are multiplied to obtain the Kalman gain coefficient.

[0057] Step S204: Extract the remaining power time curve and the remaining power from the optimal remaining power change curve based on the preset sampling length.

[0058] Among them, the sampling remaining power time curve refers to the sequence of sampling points that are traced back from the current moment and have the same sampling length; the sampling remaining power refers to the battery remaining power at each moment in the sampling remaining power time curve. The processing terminal extracts the sampling remaining power time curve from the optimal remaining power change curve based on the sampling length, and extracts all the sampling remaining power from the extracted sampling remaining power time curve.

[0059] The sampling length refers to the number of sampling points used to analyze the subsequent voltage drop trend. In one embodiment, the sampling length is 100.

[0060] Step S205: Calculate the average value of the sampled remaining power to generate the average remaining power.

[0061] The average remaining power refers to the arithmetic mean of the remaining power sampled within the sampling length. The processing terminal calculates the sum of the remaining power sampled, and then divides the sum by the sampling length to obtain the average remaining power.

[0062] Step S206: Analyze the sampling remaining power time curve and the average remaining power to generate a voltage drop trend prediction curve.

[0063] The voltage drop trend prediction curve in this step is consistent with the voltage drop trend prediction curve in step S101 above. The processing terminal can obtain the voltage drop trend prediction curve by analyzing the sampling remaining power time curve and the average remaining power. For specific methods, please refer to [reference needed]. Figure 3 The steps.

[0064] Reference Figure 3 The steps for analyzing the time curve and average remaining power of the sampled battery to generate a voltage drop trend prediction curve include: Step S300: Substitute the sampled remaining power time curve and the average remaining power into the preset remaining power decay rate calculation formula to generate the average remaining power decay rate.

[0065] The average remaining power decay rate refers to the average amount of battery power reduction per unit time. The processing terminal substitutes the sampled remaining power time curve and the average remaining power into the remaining power decay rate calculation formula to obtain the average remaining power decay rate.

[0066] The formula for calculating the remaining battery capacity decay rate is: In the formula, This refers to the average rate of decay of remaining battery power. This refers to the sampling length. It refers to the first The timestamp of each data point It refers to the average of all timestamps. This refers to the time curve of sampling remaining battery power. This refers to the average remaining power.

[0067] Step S301: Extract the latest remaining battery power from the optimal remaining power change curve.

[0068] The latest remaining battery power refers to the remaining battery power at the very end of the time axis in the optimal remaining battery power change curve. The processing terminal reads the latest remaining battery power from the optimal remaining battery power change curve to obtain the latest remaining battery power.

[0069] Step S302: Calculate the sum of the product of the average remaining power decay rate and the preset prediction period and the latest battery remaining power to generate the predicted battery remaining power.

[0070] The predicted remaining battery power refers to the remaining battery power after the prediction period, assuming the average remaining power decay rate remains unchanged. The processing terminal multiplies the average remaining power decay rate by the prediction period, and then adds the product to the latest remaining battery power to obtain the predicted remaining battery power.

[0071] The prediction period refers to the length of time covered by the predicted future changes in battery terminal voltage by the processing terminal. In one embodiment, the prediction period is 1 second.

[0072] Step S303: Analyze the sampling length and real-time discharge current to generate an average discharge current.

[0073] The average discharge current refers to the arithmetic mean of the real-time discharge current within the sampling period. First, the processing terminal extracts the sampled discharge current curve from the real-time discharge current based on the sampling length, and then extracts all the discharge currents from the extracted sampled discharge current curve. Then, the sum of all the discharge currents is calculated, and finally, the sum is divided by the sampling length to obtain the average discharge current.

[0074] Step S304: Substitute the predicted remaining battery capacity and average discharge current into the preset battery terminal voltage prediction formula to generate a voltage drop trend prediction curve.

[0075] In this step, the voltage drop trend prediction curve is consistent with the voltage drop trend prediction curve in step S101 above. The processing terminal substitutes the predicted remaining battery capacity and average discharge current into the battery terminal voltage prediction formula for calculation, and thus obtains the voltage drop trend prediction curve.

[0076] The formula for predicting battery terminal voltage is: In the formula, This refers to the voltage drop trend prediction curve. This refers to the open-circuit voltage-energy function relationship. This refers to predicting the remaining battery power. This refers to the polarization voltage variation curve. This refers to the forecast period. This refers to the polarization resistance. This refers to polarized capacitors. This refers to the average discharge current. This refers to the battery's internal resistance, which the processing terminal finds in the device's basic parameter database. For example, 1kHz.

[0077] The polarization resistor and polarization capacitor in this step are the same as those in step S200 above, and will not be repeated here.

[0078] Reference Figure 4 The steps for generating actual robot gait parameters by analyzing the voltage drop trend prediction curve and preset baseline robot gait parameters include: Step S400: Extract the lowest predicted voltage from the voltage drop trend prediction curve.

[0079] The lowest predicted voltage refers to the voltage value corresponding to the point with the smallest voltage value in the voltage drop trend prediction curve. The processing terminal extracts all voltages from the voltage drop trend prediction curve and sorts all voltage values ​​in ascending order. The first data in the sort is the lowest predicted voltage.

[0080] Step S401: Calculate the difference between the preset battery high-level threshold and the lowest predicted voltage to generate the battery maximum drop voltage.

[0081] Among them, the maximum battery drop voltage refers to the maximum extent by which the battery voltage drops from the battery high-level threshold. The larger the maximum battery drop voltage, the deeper the drop, and the greater the adjustment required. The processing terminal subtracts the minimum predicted voltage from the battery high-level threshold to obtain the maximum battery drop voltage.

[0082] The battery high-level threshold refers to the maximum voltage value of the battery under normal power supply conditions. The processing terminal extracts the nominal battery voltage from the device's basic parameter library and then multiplies the nominal battery voltage by 90% to obtain the battery high-level threshold.

[0083] Step S402: Calculate the quotient of the maximum drop voltage of the battery and the preset normal high and low voltage difference of the battery to generate the gait adjustment coefficient for this cycle.

[0084] The gait adjustment coefficient for this cycle refers to the correction coefficient used by the robot walking module to adjust the baseline robot gait parameters during the current prediction cycle. The processing terminal can obtain the gait adjustment coefficient for this cycle by dividing the maximum battery drop voltage by the normal high and low battery voltage difference.

[0085] The normal high and low voltage difference of a battery refers to the difference between the maximum and minimum voltage values ​​of the battery within its normal charging range. The processing terminal extracts the nominal voltage of the battery from the device's basic parameter library, multiplies the nominal voltage by 90% to obtain the battery's high-level threshold, extracts the battery's discharge termination voltage from the device's basic parameter library to obtain the safe low voltage threshold, and finally subtracts the safe low voltage threshold from the battery's high-level threshold to obtain the normal high and low voltage difference of the battery.

[0086] Step S403: Collect the gait adjustment coefficients from the previous cycle.

[0087] The gait adjustment coefficient of the previous cycle refers to the correction coefficient used by the robot walking module to adjust the baseline robot gait parameters during the previous prediction cycle. The processing terminal can obtain the gait adjustment coefficient of the previous cycle by extracting the gait adjustment parameters of the most recent prediction cycle from the historical data.

[0088] Step S404: Calculate the difference between the gait adjustment coefficient of the current cycle and the gait adjustment coefficient of the previous cycle to generate the gait adjustment coefficient difference between cycles.

[0089] The gait adjustment coefficient difference during the cycle refers to the difference between the gait adjustment coefficient of the current prediction cycle and the previous prediction cycle. The processing terminal subtracts the gait adjustment coefficient of the previous cycle from the gait adjustment coefficient of the current cycle to obtain the gait adjustment coefficient difference during the cycle.

[0090] Step S405: Substitute the difference in gait adjustment coefficients during the cycle and the gait adjustment coefficients of the previous cycle into the preset formula for calculating smooth gait adjustment parameters to generate the actual gait adjustment coefficients.

[0091] The actual gait adjustment coefficient refers to the correction coefficient used by the robot walking module to adjust the baseline robot gait parameters during the current prediction period. The processing terminal substitutes the difference in gait adjustment coefficients between periods and the gait adjustment coefficients of the previous period into the formula for calculating smooth gait adjustment parameters to obtain the actual gait adjustment coefficient.

[0092] The formula for calculating smooth gait adjustment parameters is: In the formula, This refers to the actual gait adjustment factor. This refers to the gait adjustment coefficient of the previous cycle. This refers to the difference in gait adjustment coefficients during the cycle. This refers to the maximum gait adjustment factor during the cycle, in order to For example, =0.05.

[0093] Step S406: Analyze the baseline robot gait parameters and the actual gait adjustment coefficients to generate the actual robot gait parameters.

[0094] In this step, the actual robot gait parameters are consistent with those in step S102 above. The processing terminal analyzes the reference robot gait parameters and the actual gait adjustment coefficients to obtain the actual robot gait parameters. The specific method is described in [reference needed]. Figure 5 The steps.

[0095] Reference Figure 5 The steps for generating actual robot gait parameters by analyzing the baseline robot gait parameters and the actual gait adjustment coefficients include: Step S500: Find the maximum step frequency, reference step frequency difference, maximum step length, reference step length difference, maximum support phase duration percentage and reference support phase duration percentage difference in the reference robot gait parameters.

[0096] Among them, the maximum cadence refers to the maximum number of steps the robot walking module can take per unit time when it is at its highest performance; the reference cadence difference is the difference between the maximum and minimum number of steps taken per unit time by the robot walking module; the maximum stride length is the maximum distance the robot walking module can cover in a single step; the reference stride length difference is the difference between the maximum and minimum distance the robot walking module can cover in a single step; the maximum support phase duration percentage refers to the maximum duration percentage of the supporting leg in the robot walking module within a complete walking cycle; the reference support phase duration percentage difference is the difference between the maximum and minimum duration percentages of the supporting leg in the robot walking module within a complete walking cycle. The processing terminal can obtain the maximum cadence, reference cadence difference, maximum stride length, reference stride length difference, maximum support phase duration percentage, and reference support phase duration percentage difference by searching in the reference robot gait parameters.

[0097] Step S501: Calculate the product of the actual gait adjustment coefficient with the reference step frequency difference, the reference step length difference, and the reference support phase duration ratio difference, respectively, to generate the step frequency adjustment amount, step length adjustment amount, and support phase duration ratio adjustment amount.

[0098] Among them, the step frequency adjustment amount refers to the adjustment amount of the number of steps the robot walking module takes per unit time in the current prediction period. The processing terminal multiplies the actual gait adjustment coefficient by the reference step frequency difference to obtain the step frequency adjustment amount.

[0099] The step length adjustment refers to the distance adjustment of a single step by the robot's walking module in the current prediction period. The processing terminal multiplies the actual gait adjustment coefficient by the reference step length difference to obtain the step length adjustment.

[0100] The support phase duration ratio adjustment refers to the adjustment of the duration ratio of the supporting leg in the robot walking module during a complete walking cycle in the current prediction period. The processing terminal multiplies the actual gait adjustment coefficient by the difference between the reference support phase duration ratio and the actual support phase duration ratio to obtain the support phase duration ratio adjustment.

[0101] Step S502: Calculate the difference between the maximum step frequency and the step frequency adjustment amount to generate the actual robot step frequency.

[0102] The actual robot step frequency refers to the number of steps the robot walking module takes per unit time in the current prediction period. The processing terminal subtracts the step frequency adjustment amount from the maximum step frequency to obtain the actual robot step frequency.

[0103] Step S503: Calculate the difference between the maximum step size and the step size adjustment amount to generate the actual robot step size.

[0104] The actual robot step length refers to the distance that the robot walking module can cover in a single step during the current prediction period. The processing terminal subtracts the step length adjustment amount from the maximum step length to obtain the actual robot step length.

[0105] Step S504: Calculate the difference between the maximum support phase duration percentage and the support phase duration percentage adjustment amount to generate the actual robot support phase duration percentage.

[0106] The actual robot support phase duration percentage refers to the percentage of time the supporting leg in the robot walking module is used for support in a complete walking cycle during the current prediction period. The processing terminal subtracts the support phase duration percentage adjustment amount from the maximum support phase duration percentage to obtain the actual robot support phase duration percentage.

[0107] Step S505: Summarize the actual robot step frequency, actual robot step length, and actual robot support phase duration ratio to generate actual robot gait parameters.

[0108] In this step, the actual robot gait parameters are the same as those in step S102 above. The processing terminal summarizes the actual robot step frequency, actual robot step length and actual robot support phase duration ratio to obtain the actual robot gait parameters.

[0109] Reference Figure 6 The steps for analyzing real-time joint position signals and real-time foot contact forces to generate joint current limiting include: Step S600: Calculate the real-time joint position signal based on the preset kinematic model to generate the forward direction component of the foot.

[0110] Among them, the forward direction component of the foot refers to the coordinate component of the foot of each leg in the robot walking module along the direction of travel in the robot's torso coordinate system. The processing terminal inputs the real-time joint position signal into the kinematic model for calculation, and the forward direction component of the foot can be obtained.

[0111] A kinematic model is a mathematical model used to describe the mapping relationship between joint angles and the spatial position of the foot. It mainly consists of a geometric parameter layer, a coordinate transformation layer, and a forward calculation layer. The geometric parameter layer inputs real-time joint position signals into the coordinate transformation layer. The coordinate transformation layer uses the center of mass of the robot's torso as the origin and sequentially connects the real-time joint position signals through a matrix to transfer the coordinate system from the origin to the end of the foot to obtain the pose matrix. The forward calculation layer extracts the three-dimensional coordinates from the pose matrix with the positive x-axis as the direction of travel and extracts the x-axis coordinate component from the three-dimensional coordinates to obtain the forward direction component of the foot.

[0112] Step S601: Analyze the forward direction component of the foot, the real-time foot contact force, and the preset support swing side characteristic parameters to generate the motion types of each side of the foot.

[0113] Among them, the movement type of each side of the foot refers to the type of gait phase of each leg in the robot walking module at the current moment. The movement types of each side of the foot are divided into two types: support movement type and swing movement type. The processing terminal determines whether the forward direction component of the foot is less than the basic threshold of the support phase. If it is less, it means that the foot is behind the torso, which may belong to the swing movement type. If it is determined at this time that the real-time foot contact force is not greater than the foot contact force threshold, it means that the foot does belong to the swing movement type. If the forward direction component of the foot is not less than the basic threshold of the support phase, it means that the foot is in front of the torso, which may belong to the support movement type. If it is determined at this time that the real-time foot contact force is greater than the foot contact force threshold, it means that the foot does belong to the support movement type.

[0114] The support swing side characteristic parameters refer to the set of threshold parameters used to determine whether the movement type of each side of the foot is a support movement type or a swing movement type. In one embodiment, the support swing side characteristic parameters include the support phase base threshold and the foot contact force threshold. The support phase base threshold is -2cm and the foot contact force threshold is 15% of the robot's own weight.

[0115] Support motion type refers to the motion type of each side of the foot when the legs are in front of the torso in the robot walking module; swing motion type refers to the motion type of each side of the foot when the legs are behind the torso in the robot walking module.

[0116] Step S602: Find the power distribution weight of each joint in the preset power distribution weight according to the movement type of each side of the foot.

[0117] Among them, the power allocation weight of each joint refers to the power allocation coefficient allocated to all joint drive motors in the mechanical structure of each leg. The processing terminal can obtain the power allocation weight of each joint by searching in the power allocation weight according to the movement type of each side of the foot.

[0118] Power allocation weight refers to the correspondence between the movement type of each side of the foot and the power allocation weight. In one embodiment, when the movement type of each side of the foot is a support movement type, the power allocation weight is 1, and when the movement type of each side of the foot is a swinging movement type, the power allocation weight is 0.3.

[0119] Step S603: Collect the maximum safe discharge current.

[0120] The maximum safe discharge current refers to the maximum continuous discharge current value allowed under the current battery state of charge and temperature conditions. The processing terminal finds the maximum safe discharge current in the battery remaining capacity temperature-current relationship table based on the current optimal remaining capacity and the battery temperature read from the thermistor attached to the battery.

[0121] The battery remaining capacity temperature-current relationship table refers to the relationship between the battery's state of charge (SOC) and temperature and the maximum safe discharge current. It is obtained by operators through a large number of experiments, considering at least three dimensions: SOC, battery temperature, and internal resistance increment. By conducting pulse discharge tests under different SOC, temperature, and internal resistance increments, the maximum safe current output by the battery is recorded, and a mapping table is formed by mapping the corresponding SOC, temperature, and internal resistance increments to the maximum safe current output.

[0122] Step S604: Calculate the product of the real-time battery terminal voltage and the maximum safe discharge current to generate the maximum available power.

[0123] The maximum available power refers to the upper limit of the total power that all joint drive motors are allowed to consume under the current battery charge state. The processing terminal multiplies the real-time battery terminal voltage by the maximum safe discharge current to obtain the maximum available power.

[0124] Step S605: Analyze the power allocation weights and maximum available power of each joint to generate joint current limits.

[0125] The joint current limiting in this step is consistent with the joint current limiting in step S104 above. The processing terminal analyzes the power allocation weights and maximum available power of each joint to obtain the joint current limiting. For specific methods, please refer to [link / reference]. Figure 7 The steps.

[0126] Reference Figure 7 The steps for analyzing the power allocation weights and maximum available power of each joint to generate joint current limits include: Step S700: Calculate the sum of the power allocation weights for each joint to generate the total power allocation weights.

[0127] The total power allocation weight refers to the sum of the power allocation coefficients allocated to all joint drive motors in the mechanical structure of each leg. The processing terminal adds up the power allocation weights of each joint to obtain the total power allocation weight.

[0128] Step S701: Calculate the quotient of the power allocation weight of each joint to the total power allocation weight, so as to generate the power allocation ratio of each joint.

[0129] The power allocation ratio of each joint refers to the proportion of the power allocation coefficient of all joint drive motors in the mechanical structure of each leg in the total power allocation weight. The processing terminal divides the power allocation weight of each joint by the total power allocation weight to obtain the power allocation ratio of each joint.

[0130] Step S702: Calculate the product of the power allocation ratio of each joint and the maximum available power to generate the power allocation for each joint.

[0131] The power allocation for each joint refers to the power distributed to all joint drive motors in the mechanical structure of each leg. The processing terminal multiplies the maximum available power by the power allocation ratio of each joint to obtain the power allocation for each joint.

[0132] Step S703: Calculate the quotient of the power allocated to each joint and the real-time battery terminal voltage to generate the joint current limit.

[0133] In this step, the joint current limit is the same as that in step S104 above. The processing terminal divides the power allocated to each joint by the real-time battery terminal voltage to obtain the joint current limit.

[0134] Based on the same inventive concept, embodiments of this application provide a voltage feedback gait adjustment system for a robot walking module, comprising: The data acquisition module is used to acquire real-time discharge current, real-time battery terminal voltage, real-time joint position signal, real-time foot contact force, gait adjustment coefficient of the previous cycle, and maximum safe discharge current. The memory is used to store the program for a voltage feedback gait adjustment method for a robot walking module; The processor can load and execute programs in memory to implement a voltage feedback gait adjustment method for a robot walking module.

[0135] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0136] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a voltage feedback gait adjustment method for a robot walking module.

[0137] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0138] Based on the same inventive concept, embodiments of this application provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to provide a voltage feedback gait adjustment method for a robot walking module.

[0139] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0140] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for voltage feedback gait adjustment in a robot walking module, characterized in that, include: Collect the real-time discharge current and real-time battery terminal voltage of the preset robot walking module; The real-time discharge current and real-time battery terminal voltage are analyzed to generate a voltage drop trend prediction curve. The voltage drop trend prediction curve and the preset baseline robot gait parameters are analyzed to generate the actual robot gait parameters. The robot's walking module is controlled to walk based on the actual robot gait parameters, and real-time joint position signals and real-time foot contact force are collected. The real-time joint position signal and real-time foot contact force are analyzed to generate joint current limiting. Power the robot's walking module according to the joint current limit, and continue to control the robot's walking module to walk according to the actual robot gait parameters; The steps for generating actual robot gait parameters by analyzing the voltage drop trend prediction curve and preset baseline robot gait parameters include: Extract the lowest predicted voltage from the voltage drop trend prediction curve; Calculate the difference between the preset battery high-level threshold and the lowest predicted voltage to generate the battery's maximum drop voltage; Calculate the quotient of the battery's maximum voltage drop and the preset difference between the normal battery's high and low voltages to generate the cycle gait adjustment coefficient. Collect the gait adjustment coefficient from the previous cycle; Calculate the difference between the gait adjustment coefficient of the current cycle and the gait adjustment coefficient of the previous cycle to generate the gait adjustment coefficient difference between cycles; The difference in gait adjustment coefficients during the cycle and the gait adjustment coefficients of the previous cycle are substituted into the preset formula for calculating smooth gait adjustment parameters to generate the actual gait adjustment coefficients. The baseline robot gait parameters and actual gait adjustment coefficients are analyzed to generate the actual robot gait parameters; The steps for generating actual robot gait parameters by analyzing the baseline robot gait parameters and the actual gait adjustment coefficients include: Find the maximum step frequency, the reference step frequency difference, the maximum step length, the reference step length difference, the maximum support phase duration percentage, and the reference support phase duration percentage difference in the reference robot gait parameters; Calculate the product of the actual gait adjustment coefficient with the reference step frequency difference, reference step length difference, and reference support phase duration percentage difference to generate the step frequency adjustment amount, step length adjustment amount, and support phase duration percentage adjustment amount; Calculate the difference between the maximum step frequency and the step frequency adjustment amount to generate the actual robot step frequency; Calculate the difference between the maximum step size and the step size adjustment to generate the actual robot step size; Calculate the difference between the maximum support phase duration percentage and the adjustment amount of the support phase duration percentage to generate the actual robot support phase duration percentage; The actual robot's step frequency, actual robot step length, and actual robot support phase duration are summarized to generate actual robot gait parameters.

2. The method for voltage feedback gait adjustment of a robot walking module according to claim 1, characterized in that, The steps for analyzing real-time discharge current and real-time battery terminal voltage to generate a voltage drop trend prediction curve include: The real-time discharge current is input into the preset state equation for calculation to generate the uncorrected remaining charge curve and polarization voltage change curve. The uncorrected remaining charge curve and polarization voltage change curve are input into the preset observation equation for calculation to generate the theoretical battery terminal voltage. Calculate the difference between the real-time battery terminal voltage and the theoretical battery terminal voltage to generate the battery terminal voltage residual; The uncorrected remaining power curve is corrected based on the preset Kalman gain coefficient and battery terminal voltage residual to generate the optimal remaining power change curve. Based on the preset sampling length, the remaining power time curve and the remaining power are extracted from the optimal remaining power change curve. Calculate the average of the sampled remaining battery power to generate the average remaining battery power; The sampling remaining power time curve and average remaining power are analyzed to generate a voltage drop trend prediction curve.

3. The method for voltage feedback gait adjustment of a robot walking module according to claim 2, characterized in that, The steps for analyzing the time curve of the sampled remaining power and the average remaining power to generate a voltage drop trend prediction curve include: The sampled remaining power time curve and average remaining power are substituted into the preset remaining power decay rate calculation formula to generate the average remaining power decay rate. Extract the latest remaining battery capacity from the optimal remaining capacity change curve; The product of the average remaining battery capacity decay rate and the preset prediction period is summed with the latest remaining battery capacity to generate the predicted remaining battery capacity. The sampling length and real-time discharge current are analyzed to generate an average discharge current; The predicted remaining battery capacity and average discharge current are substituted into a preset battery terminal voltage prediction formula to generate a voltage drop trend prediction curve.

4. The method for voltage feedback gait adjustment of a robot walking module according to claim 1, characterized in that, The steps for analyzing real-time joint position signals and real-time foot contact forces to generate joint current limiting include: The real-time joint position signal is calculated based on a preset kinematic model to generate the forward direction component of the foot. The forward direction component of the foot, the real-time contact force of the foot, and the preset support swing side characteristic parameters are analyzed to generate the motion types of each side of the foot. The power allocation weights for each joint are determined from the preset power allocation weights based on the movement type of each side of the foot. Collect the maximum safe discharge current; Calculate the product of the real-time battery terminal voltage and the maximum safe discharge current to generate the maximum usable power; The power allocation weights and maximum available power of each joint are analyzed to generate joint current limits.

5. The method for voltage feedback gait adjustment of a robot walking module according to claim 4, characterized in that, The steps for analyzing the power allocation weights and maximum available power of each joint to generate joint current limits include: Calculate the sum of the power allocation weights for each joint to generate the total power allocation weights; Calculate the quotient of the power allocation weight of each joint to the total power allocation weight to generate the power allocation ratio of each joint; Calculate the product of the power allocation ratio of each joint and the maximum available power to generate the power allocation for each joint; Calculate the quotient of the power allocated to each joint and the real-time battery terminal voltage to generate the joint current limit.

6. A voltage feedback gait adjustment system for a robot walking module, characterized in that, include: The acquisition module is used to acquire real-time discharge current, real-time battery terminal voltage, real-time joint position signal, and real-time foot contact force. A memory for storing a program for a robot walking module voltage feedback gait adjustment method as described in any one of claims 1 to 5; The processor and the program in the memory can be loaded and executed by the processor to implement the voltage feedback gait adjustment method for a robot walking module as described in any one of claims 1 to 5.

Images