A control method for robotaxi drive motor degradation perception and autonomous maintenance

Abstract

The application relates to a control method for Robotaxi driving motor degradation sensing and autonomous maintenance, comprising the following steps: S1, constructing a nonlinear degradation model of a Robotaxi driving motor to obtain quantitative indexes of the remaining service life and reliability of the driving motor; S2, constructing a travel task risk evaluation function according to the remaining service life of the driving motor to obtain a quantitative value of Robotaxi travel risk; S3, based on the reliability in step S1 and the risk quantitative value in step S2, designing a self-tuning model predictive control strategy, applying the solved optimal control quantity to the Robotaxi, and realizing autonomous maintenance of the Robotaxi driving motor. Compared with the prior art, the application comprehensively considers the health state of the driving motor and the task demand of the Robotaxi, and designs a closed-loop controller containing the implicit degradation amount of the driving motor to prolong the operation cycle of the Robotaxi, which has great significance for realizing the satisfactory balance of the control performance and reliability of the Robotaxi.

Description

Technical Field

[0001] This invention relates to the field of automotive control, and in particular to a control method for detecting and autonomously maintaining the degradation of Robotaxi drive motors. Background Technology

[0002] Robotaxis, a product of advanced autonomous driving technology, help alleviate traffic congestion, save travel costs, reduce vehicle usage, and reallocate public space to improve the quality of urban life. However, unlike other unmanned systems operating under fixed conditions, Robotaxis need to execute randomly booked rides from consumers. In variable operating scenarios, failure modes of critical components can easily affect the reliability and safety of Robotaxis. State-based maintenance is a proactive maintenance strategy that executes maintenance plans based on the system's current health status. Prognostic and health management technologies, as enabling technologies for state-based maintenance, provide system-level support for its development. If prognostic and health management technologies are embedded into Robotaxis to achieve autonomous maintenance, the reliability and safety issues of Robotaxis are expected to be resolved.

[0003] The drive motor is one of the key components of a Robotaxi. The health of the drive motor directly determines the Robotaxi's operational risk. Using prognostic and management technologies to achieve drive motor degradation detection and autonomous maintenance has significant practical implications and application prospects. Traditionally, closed-loop control algorithms do not consider the health state of the drive motor. When the drive motor's performance significantly degrades, the controller output increases to achieve the desired dynamic performance. Considering the damage accumulation and degradation dynamics of electromechanical actuators, the control force is proportional to the rate of degradation of the drive motor; larger control actions will accelerate its degradation and failure process. Therefore, there is an urgent need to develop a control algorithm that allows Robotaxis to achieve better closed-loop performance while minimizing the cumulative degradation of its drive motor.

[0004] Therefore, designing and developing a control method for Robotaxi drive motor degradation sensing and autonomous maintenance is of great significance for extending the operating cycle of Robotaxi and achieving a satisfactory balance between control performance and reliability. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a control method for Robotaxi drive motor degradation sensing and autonomous maintenance. By comprehensively considering the health status of the drive and the requirements of Robotaxi tasks, a closed-loop controller containing the implicit degradation of the drive motor is designed to extend the Robotaxi operating cycle and achieve a satisfactory balance between Robotaxi control performance and reliability.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A control method for Robotaxi drive motor degradation sensing and autonomous maintenance includes the following steps:

[0008] S1. Construct a nonlinear degradation model for the Robotaxi drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor;

[0009] S2. Based on the remaining service life of the drive motor, construct a travel task risk assessment function to obtain the Robotaxi travel risk quantification value.

[0010] S3. Based on the reliability in step S1 and the risk quantification value in step S2, design a self-tuning model predictive control strategy, apply the solved optimal control quantity to Robotaxi, and realize the autonomous maintenance of Robotaxi drive motor.

[0011] Further, step S1 includes the following sub-steps:

[0012] S101. Based on the degradation mechanism of the drive motor, a nonlinear degradation model of the drive motor is established using the nonlinear Wiener stochastic process;

[0013] S102. Based on the nonlinear degradation model of the drive motor, construct the probability density function and reliability function of the remaining service life of the drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor.

[0014] Further, in step S1, the expression for the nonlinear degradation model of the drive motor is:

[0015]

[0016] In the formula, x0 is the initial degradation of the motor, μ(τ,θ) is the drift coefficient of the degradation rate of the motor, σ is a constant diffusion coefficient, B(t) is the standard Brownian motion, which reflects the random dynamics in the degradation process, and B(t)~N(0,t).

[0017] Wherein, the drift coefficient μ(τ,θ) of the motor degradation rate is a nonlinear function, and its expression is:

[0018] μ(τ,θ)=μ0exp(κ|u(τ)|)

[0019] In the formula, μ0 and κ are correlation coefficients, u(τ) is the control force acting on the motor, and |·| is the absolute value of the parameter.

[0020] Further, in step S102, the probability density function of the remaining service life of the drive motor is:

[0021]

[0022] In the formula,

[0023]

[0024] ω is the degradation threshold of the drive motor;

[0025] The reliability function for the remaining service life of the drive motor is:

[0026]

[0027] In the formula, Φ(·) is the distribution function of the standard normal distribution;

[0028] Based on the state and degradation threshold of the drive motor at different times, the probability density function and reliability function of the remaining service life are solved to obtain quantitative indicators of the remaining service life and reliability of the drive motor.

[0029] Further, step S2 includes the following sub-steps:

[0030] S201. Using the remaining service life and expected life extension time of the drive motor in step S1 as independent variables, construct a Robotaxi travel mission risk assessment function.

[0031] S202. Based on the state of the drive motor at different times, solve the risk assessment function to obtain the quantitative value of Robotaxi travel risk.

[0032] Further, in step S201, the expression for the risk assessment function is:

[0033]

[0034] In the formula, χ1, χ2, β1, β2, ξ1 and ξ2 are correlation coefficients, X(t) is the nonlinear degradation model of the drive motor, and t el The desired lifespan extension;

[0035] In step S202, based on the quantitative value of the remaining service life X(t) of the drive motor obtained in step S1 and the expected life extension time, the risk assessment function is solved to obtain the Robotaxi travel risk value at different times.

[0036] The expected life extension time t el The expression is:

[0037] t el =t task -t life

[0038] In the formula, t taskFor the remaining time of the travel task, t life This refers to the remaining service life of the drive motor.

[0039] Furthermore, step S3 includes the following sub-steps:

[0040] S301. Construct a longitudinal dynamic model of Robotaxi and incorporate the reliability of the drive motor from step S1 into the state space equation of the system to establish a state space model that considers the reliability of the drive motor.

[0041] S302. Based on the Robotaxi travel risk quantification value in step S2, design a self-tuning model predictive control strategy and solve for the optimal control quantity applied to the vehicle.

[0042] Further, in step S301, the Robotaxi longitudinal dynamics model is modeled as a first-order inertial element, expressed as:

[0043]

[0044] In the formula, A∈R n×n Let B be the state matrix of the system, and B ∈ R. n×1 Let C be the input matrix of the system, and R ∈ R. 1×n Let x(t) ∈ R be the output matrix of the system. n×1 Let u(t) be the system's state vector, u(t) be the system's control input, and y(t) be the system's output.

[0045] Incorporating the reliability of the drive motor into the system's state-space equations, the state-space model expression considering the drive motor's reliability is established as follows:

[0046]

[0047] Further, in step S302, the prediction time domain of the self-tuning model predictive control strategy is defined as p, and the control time domain is defined as m. Then, the output Y of the system in the next p steps is... p The expression for (k+1|k) is:

[0048] Y p (k+1|k)=ψx(k)+θΔU(k)

[0049] In the formula, the expression for θ is:

[0050]

[0051] The objective function of the self-tuning model predictive control strategy is defined as follows:

[0052]

[0053] In the formula, the output weight matrix Γy =diag(Γ y,1 ,Γ y,2 ,...Γ y,p ), Y ref For outputting the reference trajectory, k1 is the torque coefficient of the drive motor, k2 and k3 are the coefficients of copper loss and iron loss of the drive motor, respectively, I is the coil current of the drive motor, n is the speed of the drive motor, k is the discrete time step, and T is the torque coefficient of the drive motor. m This refers to the torque of the drive motor.

[0054] Furthermore, the constraints of the self-tuning model predictive control strategy system are defined as follows:

[0055]

[0056] Δu min ≤Δu(k)≤Δu max

[0057] u min ≤u(k)≤u max

[0058] y min ≤y(k)≤y max

[0059] Δu(k) is the input increment, Δu min Δu is the minimum value of the input increment. max u is the maximum value of the input increment. min u is the minimum value of the input quantity. max y represents the maximum value of the input quantity. min The minimum value of the output, y max This represents the maximum value of the output.

[0060] By solving the self-tuning model predictive control strategy at different times and applying the obtained optimal control quantity to the vehicle, Robotaxi can achieve autonomous maintenance under the implicit degradation of the drive motor, extend the Robotaxi's operating cycle, and achieve a satisfactory balance between Robotaxi control performance and reliability.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] This invention, based on a nonlinear Wiener stochastic process, predicts the remaining service life of the drive motor, accurately senses its current degradation state, and obtains a quantified reliability value. Subsequently, this invention proposes for the first time a risk assessment function to quantify the risks in travel tasks and use it as input to the self-tuning model predictive control. Finally, this invention innovatively incorporates the drive motor reliability and the quantified travel risk values ​​into the self-tuning model predictive control, achieving autonomous maintenance of Robotaxi. Research shows that this control method for Robotaxi drive motor degradation sensing and autonomous maintenance is reasonable, feasible, and effective, and is of great significance for achieving a satisfactory balance between extending Robotaxi operating cycles and Robotaxi control performance and reliability. Attached Figure Description

[0063] Figure 1 This is a block diagram of a control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to the present invention;

[0064] Figure 2 This is a graph showing the changes in risk quantification value and the weight matrix of the self-tuning model predictive control strategy in an embodiment of the present invention.

[0065] Figure 3 The simulation results of vehicle speed and drive motor torque under the action of the risk assessment function in the embodiments of the present invention are shown.

[0066] Figure 4 This is a graph showing the degradation process of the drive motor before and after the application of the risk assessment function in an embodiment of the present invention.

[0067] Figure 5 This is a comparison chart of the remaining lifespan of the drive motor before and after the application of the risk assessment function in an embodiment of the present invention. Detailed Implementation

[0068] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0069] like Figure 1 As shown, this invention provides a control method for Robotaxi drive motor degradation sensing and autonomous maintenance, including the following steps:

[0070] 1) Construct a nonlinear degradation model for the Robotaxi drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor;

[0071] 11) Based on the degradation mechanism of the drive motor, a nonlinear degradation model of the drive motor is established using a nonlinear Wiener stochastic process;

[0072] 12) Based on the nonlinear degradation model, construct the probability density function and reliability function of the remaining service life of the drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor.

[0073] 2) Based on the remaining service life of the drive motor, construct a travel task risk assessment function to obtain the quantitative value of Robotaxi travel risk;

[0074] 21) Using the remaining service life and expected life extension time of the drive motor in step 1) as independent variables, construct a risk assessment function for Robotaxi travel missions.

[0075] 22) Based on the state of the drive motor at different times, solve the risk assessment function to obtain the quantitative value of Robotaxi travel risk.

[0076] 3) Based on the reliability in step 1) and the risk quantification value in step 2), design a self-tuning model predictive control strategy, apply the solved optimal control quantity to Robotaxi, and realize the autonomous maintenance of Robotaxi drive motor.

[0077] 31) Construct a longitudinal dynamic model of Robotaxi and incorporate the reliability of the drive motor from step 1) into the state-space equations of the system to establish a state-space model that considers the reliability of the drive motor.

[0078] 32) Based on the Robotaxi travel risk quantification value in step 2), design a self-tuning model predictive control strategy and solve for the optimal control quantity applied to the vehicle.

[0079] In this embodiment, the specific steps for Robotaxi drive motor degradation detection and autonomous maintenance implemented according to the above method are as follows:

[0080] Step 1: Establish a nonlinear degradation model for the drive motor using a nonlinear Wiener stochastic process. Its expression is:

[0081]

[0082] In the formula, x0 is the initial degradation of the motor, μ(τ,θ) is the drift coefficient of the motor degradation rate, which is a nonlinear function. σ is a constant diffusion coefficient, and B(t) is the standard Brownian motion, reflecting the random dynamics in the degradation process, and B(t)~N(0,t).

[0083] Furthermore, by relating the control force of the drive motor to its degradation, the drift coefficient in the above equation is constructed as follows:

[0084] μ(τ,θ)=μ0exp(κ|u(τ)|)

[0085] In the formula, μ0 and κ are correlation coefficients, u(τ) is the control force acting on the motor, and |·| is the absolute value of the parameter.

[0086] Furthermore, considering the measurement uncertainty in the degradation model, the relationship between the monitored degradation value and the true value is defined as follows:

[0087] Y(t)=X(t)+ε

[0088] In the formula, ε represents the random measurement error, which is assumed to follow the same normal distribution at different monitoring times, i.e., ε ~ N(0, γ). 2 ), γ 2 The variance is the variance of the normal distribution.

[0089] Furthermore, the approximate expression for the probability density function of the remaining useful life in closed-form is:

[0090]

[0091] In the formula,

[0092]

[0093] ω is the degradation threshold of the drive motor;

[0094] Furthermore, the reliability of the drive motor is defined as the probability that the current degradation amount is less than the degradation threshold, expressed as:

[0095]

[0096] In the formula, Φ(·) is the distribution function of the standard normal distribution.

[0097] After obtaining the probability density function and reliability function of the remaining service life of the drive motor, the quantitative indicators of the remaining service life and reliability of the drive motor are obtained by solving.

[0098] The parameters of the above-mentioned drive motor remaining life prediction model are shown in Table 1:

[0099] Table 1

[0100] <![CDATA[μ0]]> Degradation rate 0.0227 κ Control force coefficient 0.01 σ diffusion coefficient 0.0011 <![CDATA[x0]]> Initial degradation 0

[0101] Step 2: Using the remaining service life and expected life extension time of the drive motor from Step 1 as independent variables, construct a Robotaxi travel mission risk assessment function to quantify the risks of Robotaxi in travel missions.

[0102] The expression for the expected life extension time is defined as follows:

[0103] t el =t task -tlife

[0104] In the formula, t task For the remaining time of the travel task, t life This refers to the remaining service life of the drive motor.

[0105] Furthermore, the current degradation X(t) of the drive motor and the expected life extension t are considered. el As a risk variable, the risk assessment function expression is constructed as follows:

[0106]

[0107] In the formula, χ1, χ2, β1, β2, ξ1 and ξ2 are correlation coefficients.

[0108] like Figure 2 As shown, the risk value J increases with the increase in Robotaxi task execution time. risk The risk value increases in an "S"-shaped curve. In the early stages of the mission, the risk value is low, and there is a certain margin between the degradation of the drive motor and the failure threshold, meaning the Robotaxi is likely to complete the trip safely. As the mission progresses, the health of the drive motor deteriorates, the risk value rises rapidly and remains high, indicating that the Robotaxi's condition is insufficient to complete the mission and it may stop en route due to drive motor failure. Furthermore, the input and output weight matrices of the self-tuning model predictive controller are adjusted according to the changes in the risk value.

[0109] Step 3: Based on the reliability in Step 1 and the risk quantification value in Step 2, design a self-tuning model predictive control strategy, apply the solved optimal control quantity to Robotaxi, and realize the autonomous maintenance of Robotaxi drive motor.

[0110] The longitudinal dynamics of the Robotaxi are modeled as a first-order inertial element, expressed as:

[0111]

[0112] In the formula, A∈R n×n , B∈R n×1 and C∈R 1×n Let x(t) ∈ R be the system's state matrix, input matrix, and output matrix, respectively. n×1 Let x be the state vector of the system, and in this invention, we take x = [v, a]. T v is the vehicle speed, a is the vehicle acceleration, and u(t) is the system control input, which is taken as u(t) = T in this invention. m T m Let y(t) be the torque of the driving motor. y(t) is the output of the system, which is taken as y(t) = v in this invention.

[0113] Furthermore, the degradation of the reconstructed drive motor is:

[0114] X(t) = z(t) + σB(t)

[0115] Furthermore, the reliability of the drive motor is incorporated into the system's state-space equations, establishing a state-space model that considers the reliability of the drive motor, expressed as:

[0116]

[0117] Furthermore, defining the prediction time domain of the self-tuning model predictive control as p and the control time domain as m, then the output Y of the system in the next p steps is... p (k+1|k) is represented as:

[0118] Y p (k+1|k)=ψx(k)+θΔU(k)

[0119] In the formula, the expression for θ is:

[0120]

[0121] Furthermore, the objective function of the self-tuning model predictive controller in the Robotaxi speed tracking problem is defined as follows:

[0122]

[0123] In the formula, the output weight matrix Γ y =diag(Γ y,1 ,Γ y,2 ,...Γ y,p ), Y ref For outputting the reference trajectory, k1 is the torque coefficient of the drive motor, k2 and k3 are the coefficients of copper loss and iron loss of the drive motor, respectively, I is the coil current of the drive motor, n is the speed of the drive motor, k is the discrete time step, and T is the torque coefficient of the drive motor. m This refers to the torque of the drive motor.

[0124] Furthermore, the system's constraints are defined as follows:

[0125]

[0126] Δu min ≤Δu(k)≤Δu max

[0127] u min ≤u(k)≤u max

[0128] y min ≤y(k)≤ymax

[0129] The predictive controller parameters of the self-tuning model mentioned above are shown in Table 2:

[0130] Table 2

[0131] <![CDATA[T s ]]> Sampling time 0.02 p Prediction Time Domain 15 m Control Time Domain 5 <![CDATA[Γ y ]]> Output weight matrix initial values 100 <![CDATA[Γ u ]]> Output weight matrix initial values 10

[0132] The self-tuning model predictive control strategy is solved at different times, and the obtained optimal control quantity is applied to the vehicle.

[0133] The algorithm of this invention will be simulated and verified in a Robotaxi speed tracking example:

[0134] Figure 3 The simulation results of Robotaxi speed and drive motor torque under the influence of the risk assessment function are presented. Figure 2 As shown in Figure 3, it can be seen that the risk assessment function further suppresses the control quantity by changing the weight matrix of the predictive control of the self-tuning model, so that the drive motor outputs a smaller torque value to slow down its degradation process, thereby achieving autonomous maintenance. Figure 4 The degradation process of the drive motor before and after the application of the risk assessment function was compared. During the stage of rapid increase in risk value, the degradation rate of the drive motor under the model predictive controller considering the risk assessment function gradually slowed down. Figure 5 The simulation results demonstrate that the use of the risk assessment function can appropriately extend the lifespan of the drive motor and enable autonomous maintenance of Robotaxi.

[0135] The simulation results show that the present invention can effectively realize the degradation detection and autonomous maintenance of Robotaxi drive motors, which is of great significance for promoting the road application of Robotaxi.

[0136] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A control method for Robotaxi drive motor degradation sensing and autonomous maintenance, characterized in that, Includes the following steps: S1. Construct a nonlinear degradation model for the Robotaxi drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor; S2. Based on the remaining service life of the drive motor, construct a travel task risk assessment function to obtain the Robotaxi travel risk quantification value. S3. Based on the reliability in step S1 and the risk quantification value in step S2, design a self-tuning model predictive control strategy, apply the solved optimal control quantity to Robotaxi, and realize the autonomous maintenance of Robotaxi drive motor. Step S2 includes the following sub-steps: S201. Using the remaining service life and expected life extension time of the drive motor in step S1 as independent variables, construct a Robotaxi travel mission risk assessment function. S202. Based on the state of the drive motor at different times, solve the risk assessment function to obtain the quantitative value of Robotaxi travel risk; In step S201, the expression for the risk assessment function is: In the formula, , , , , and The correlation coefficient is... For the nonlinear degradation model of the drive motor, The desired lifespan extension; The expected life extension time The expression is: In the formula, Remaining time for the trip. This refers to the remaining service life of the drive motor.

2. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 1, characterized in that, Step S1 includes the following sub-steps: S101. Based on the degradation mechanism of the drive motor, a nonlinear degradation model of the drive motor is established using the nonlinear Wiener stochastic process; S102. Based on the nonlinear degradation model of the drive motor, construct the probability density function and reliability function of the remaining service life of the drive motor to obtain quantitative indicators of the remaining service life and reliability of the drive motor.

3. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 1, characterized in that, In step S1, the expression for the nonlinear degradation model of the drive motor is: In the formula, This represents the initial degradation of the motor. The drift coefficient is the rate of motor degradation. For a constant diffusion coefficient, It is standard Brownian motion, reflecting the stochastic dynamics of the degradation process, and ; Wherein, the drift coefficient of the motor degradation rate It is a nonlinear function, and its expression is: In the formula, and The correlation coefficient is... The control force acting on the motor, This is to take the absolute value of the parameter.

4. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 3, characterized in that, In step S102, the probability density function of the remaining service life of the drive motor is: In the formula, This is the degradation threshold for the drive motor; The reliability function for the remaining service life of the drive motor is: In the formula, The distribution function of the standard normal distribution; Based on the state and degradation threshold of the drive motor at different times, the probability density function and reliability function of the remaining service life are solved to obtain quantitative indicators of the remaining service life and reliability of the drive motor.

5. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 4, characterized in that, Step S3 includes the following sub-steps: S301. Construct a longitudinal dynamic model of Robotaxi and incorporate the reliability of the drive motor from step S1 into the state space equation of the system to establish a state space model that considers the reliability of the drive motor. S302. Based on the Robotaxi travel risk quantification value in step S2, design a self-tuning model predictive control strategy and solve for the optimal control quantity applied to the vehicle.

6. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 5, characterized in that, In step S301, the Robotaxi longitudinal dynamics model is modeled as a first-order inertial element, and its expression is: In the formula, Let be the state matrix of the system. The input matrix of the system, The output matrix of the system, Let be the state vector of the system. For the system's control input, This is the system output; Incorporating the reliability of the drive motor into the system's state-space equations, the state-space model expression considering the drive motor's reliability is established as follows: 。 7. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 6, characterized in that, In step S302, the prediction time domain of the self-tuning model predictive control strategy is defined as follows: Control time domain is Then the system in the future Step output The expression is: In the formula, The expression is: The objective function of the self-tuning model predictive control strategy is defined as follows: In the formula, the output weight matrix , To output the reference trajectory, This is the torque coefficient of the drive motor. , These are the coefficients for copper loss and iron loss of the drive motor, respectively. To drive the motor speed, For discrete time steps, This refers to the torque of the drive motor.

8. The control method for Robotaxi drive motor degradation sensing and autonomous maintenance according to claim 6, characterized in that, The constraints of the self-tuning model predictive control strategy system are defined as follows: For input increment, The minimum value of the input increment. The maximum value of the input increment. The minimum value of the input quantity. The maximum value of the input quantity. The minimum value of the output quantity. This represents the maximum value of the output.

Images