Torque initialization and adaptive active-disturbance-rejection control method for constant-speed cruise switching

Abstract

The application provides a torque initialization and adaptive active disturbance rejection control method for constant speed cruise switching, and relates to the technical field of constant speed cruise control, and the method comprises the following steps: acquiring an actual execution torque of a power source at a time point before a vehicle enters a constant speed cruise mode; querying a cruise basic compensation torque table according to a current actual vehicle speed to obtain a cruise basic compensation torque; using the actual execution torque of the power source at the time point before entering the cruise to initialize a state of an extended state observer in a fuzzy adaptive linear active disturbance rejection controller; generating an original cruise torque request; performing amplitude limiting processing on the original cruise torque request according to a maximum driving torque and a maximum energy recovery torque allowed by a motor at present; and updating the extended state observer by using an actual execution torque fed back by the power source. The method can solve the torque mutation problem during constant speed cruise switching, and can realize torque smooth connection, vehicle speed stable tracking and strong anti-interference capability.

Description

Technical Field

[0001] This invention relates to the field of cruise control technology, specifically to a torque initialization and adaptive active disturbance rejection control method for cruise control switching. Background Technology

[0002] Cruise control is a common function in the longitudinal control system of modern automobiles. Existing vehicle cruise control methods typically employ proportional-integral-derivative (PID) closed-loop control based on vehicle speed deviation, combined with vehicle driving resistance compensation, acceleration feedforward, or segmented calibration tables to calculate the target torque. At the same time, to improve torque fluctuation, post-processing measures such as torque limiting, rate of change limiting, or low-pass filtering are often added to the controller output.

[0003] However, the aforementioned existing technologies have the following technical problems at the moment when the vehicle switches from normal driving mode, coasting energy recovery mode, or other longitudinal control modes to cruise control mode: the initial output of the cruise controller is usually set to zero torque or a recalculated closed-loop torque, which fails to maintain continuity with the actual torque executed by the vehicle at the moment before entering cruise, resulting in sudden torque changes and longitudinal jerking at the moment of entry; the integral state and observer state inside the controller are inconsistent with the actual longitudinal dynamic state of the vehicle, and even if output smoothing is performed, it is easy to cause response lag or repeated corrections; in addition, there is a deviation between the original torque request used by the controller and the torque actually executed by the power system due to torque limiting and other reasons, causing a mismatch between the internal estimated state and the actual force state of the vehicle, affecting the stability of vehicle speed tracking. Summary of the Invention

[0004] To address the problems in related technologies, this invention provides a torque initialization and adaptive active disturbance rejection control method for cruise control switching.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] The torque initialization and adaptive disturbance rejection control method for cruise control switching includes the following steps:

[0007] Obtain the actual torque of the power source at the moment before the vehicle enters cruise control mode;

[0008] Based on the current actual vehicle speed, consult the cruise basic compensation torque table to obtain the cruise basic compensation torque;

[0009] The extended state observer in the fuzzy adaptive linear active disturbance rejection controller is initialized using the actual torque of the power source at the moment before entering cruise.

[0010] The cruise base compensation torque is added to the torque output by the fuzzy adaptive linear active disturbance rejection controller to generate the original cruise torque request.

[0011] Based on the current maximum allowable driving torque and maximum energy recovery torque of the motor, the original cruise torque request is limited to obtain a limited cruise torque request.

[0012] The expansion state observer is updated using the actual execution torque fed back from the power source.

[0013] Optionally, the step of obtaining the actual torque of the power source at the moment before the vehicle enters cruise control mode further includes:

[0014] In the control cycle preceding the vehicle's switch from the current driving mode to cruise control mode, the vehicle controller records the actual torque executed by the power source and denotes it as:

[0015]

[0016] In the formula, This refers to the actual torque output of the powertrain that participates in the longitudinal control of the vehicle just before it enters cruise control. The actual torque fed back by the power source in the control cycle before entering cruise.

[0017] Optionally, the step of querying the cruise basic compensation torque table based on the current actual vehicle speed further includes:

[0018] The vehicle's actual speed is obtained the instant it enters cruise control mode. and according to Query the pre-calibrated vehicle speed-cruise baseline compensation torque table to obtain the cruise baseline compensation torque corresponding to the current vehicle speed. , is represented as:

[0019]

[0020] In the formula, This is a lookup function for the pre-calibrated vehicle speed-cruise baseline compensation torque.

[0021] Optionally, the step of initializing the state of the extended state observer further includes:

[0022] Initialize the vehicle speed observation state of the extended state observer to the actual vehicle speed at the moment before entering cruise:

[0023]

[0024] The total disturbance observation state of the extended state observer is initialized based on the actual torque of the power source at the moment before entering cruise:

[0025]

[0026] In the formula, This is the initial value for the vehicle speed observation state. The actual vehicle speed at the moment cruise control is activated. The initial values ​​for the total disturbance observation state, This is the nominal longitudinal dynamics input gain estimate used when the vehicle enters cruise control. The actual torque delivered by the power source just before entering cruise control.

[0027] Optionally, it also includes the step of correcting the parameters of the linear active disturbance rejection controller online using a fuzzy controller:

[0028] Based on vehicle speed error and vehicle speed error change rate The disturbance compensation coefficient correction amount is output through the fuzzy controller. , bandwidth correction of extended state observer and controller bandwidth correction And calculate the corrected parameters:

[0029]

[0030] In the formula, , and These are the basic calibration values ​​for the disturbance compensation coefficient, the extended state observer bandwidth, and the controller bandwidth, respectively. , , These represent the disturbance compensation coefficient corrected in the k-th control cycle, the extended state observer bandwidth, and the controller bandwidth, respectively, and the vehicle speed error. Target speed With actual vehicle speed The difference, the rate of change of the vehicle speed error The difference between the current cycle speed error and the previous cycle speed error is used to determine the control cycle. The ratio is calculated.

[0031] Optionally, before calculating the vehicle speed error, the method further includes a step of smoothing the target vehicle speed for cruise control using a tracking differentiator.

[0032] The target vehicle speed is obtained by using a discrete form of a tracking differentiator. Processing is performed to obtain a smooth set speed. and its rate of change Specifically:

[0033]

[0034] In the formula, For the first The cruise control sets the vehicle speed for each control cycle. To smoothly set the vehicle speed to track the output of the differentiator, To smoothly set the rate of change of vehicle speed, To control the cycle, To track the differentiator bandwidth parameter, This is to track the smooth acceleration term calculated internally by the differentiator.

[0035] Optionally, the step of the fuzzy adaptive linear active disturbance rejection controller outputting torque further includes:

[0036] Based on smooth speed setting Vehicle speed observations output by the extended state observer Calculate vehicle speed tracking error ;

[0037] The basic control quantity is calculated based on the vehicle speed tracking error. , where controller gain By the corrected controller bandwidth Determined, that is ;

[0038] The total disturbance estimated based on the basic control quantity and the extended state observer. Calculate the output torque of the fuzzy adaptive linear active disturbance rejection controller. :

[0039]

[0040] In the formula, This is the corrected disturbance compensation coefficient.

[0041] Optionally, the step of generating the original cruise torque request further includes:

[0042] The cruise base compensation torque corresponding to the current vehicle speed With fuzzy adaptive linear active disturbance rejection controller output torque Add them together to get the original cruise torque request without the limiting process. :

[0043]

[0044] Among them, in the first control cycle when the vehicle enters cruise control. The original cruise torque request Cruise baseline compensation torque at the moment of entering cruise. and the initial output torque calculated based on the state of the expanded state observer after initialization. Add them together to get the result.

[0045] Optionally, the step of limiting the original cruise torque request further includes:

[0046] The limiting process is represented as follows:

[0047]

[0048] In the formula, The limiting function is defined as follows:

[0049]

[0050] in, For the cruise torque request after limiting, The original cruise torque request without limiting processing. This represents the maximum allowable drive torque of the motor in the current control cycle, and its value must be positive. This is the maximum energy recovery torque allowed by the motor in the current control cycle, and its value is negative.

[0051] Optionally, in the step of updating the extended state observer using the actual execution torque feedback from the power source, the discrete update form of the linear extended state observer is:

[0052]

[0053] in, For observation error, For the first Vehicle speed observation values ​​for each control cycle For the first The actual vehicle speed during each control cycle To control the cycle, For the first Total disturbance observations for each control cycle For the first Disturbance compensation coefficient for each control cycle. The actual torque of the power source input to the linearly extended state observer, and , The actual torque signal fed back by the power source controller. For the observer gain of state estimation error feedback, The observer gain is used for feedback of the extended perturbation state error.

[0054] Beneficial effects:

[0055] 1. First, when the vehicle enters cruise control mode, this invention does not directly initialize the cruise controller output to zero torque, nor does it directly add the actual torque executed by the power source at the moment before entering cruise to the final output torque. Instead, it uses the actual torque executed by the power source at the moment before entering cruise to initialize the extended state observer, enabling the AFLADRC controller to output an initial control torque that matches the actual longitudinal dynamic state of the vehicle before entering cruise in the first control cycle. This avoids the controller starting from a zero state or an unreasonable state, reducing torque surges at the moment of cruise control switching.

[0056] Second, this invention obtains the cruise baseline compensation torque by looking up a table based on the current actual vehicle speed, and adds this cruise baseline compensation torque as a feedforward compensation term to the AFLADRC output torque to form the original cruise torque request. In this way, the cruise baseline compensation torque can provide the basic torque required to maintain cruise at the current vehicle speed, and the AFLADRC output torque can be corrected in real time based on vehicle speed error, error change rate, and total disturbance estimation results, thus balancing cruise entry smoothness and cruise speed tracking capability.

[0057] Third, when the vehicle switches from coasting energy recovery state or zero torque coasting state to cruise control, the present invention can initialize and inherit the actual power state before entering cruise by expanding the state observer, and weaken the negative torque drag effect by the cruise base compensation torque corresponding to the current vehicle speed, so as to avoid the longitudinal impact of deceleration and acceleration due to unreasonable initial state of the controller after the vehicle enters cruise control, thereby improving the driving comfort during the cruise control switching process.

[0058] Fourth, this invention employs AFLADRC to estimate and compensate for the total disturbance during vehicle operation. It can treat changes in vehicle mass, load, road gradient, rolling resistance, air resistance, and powertrain output capability as a unified total disturbance. Therefore, compared to fixed-parameter PI / PID control methods, this invention has lower dependence on precise vehicle dynamics models and fixed calibration parameters, and can maintain better speed stability and disturbance resistance when vehicle load, gradient, or road resistance changes.

[0059] Fifth, this invention, based on vehicle speed error and its rate of change, uses a fuzzy controller to online correct the disturbance compensation coefficient, expand the state observer bandwidth, and extend the controller bandwidth, enabling the controller to adaptively adjust the control intensity according to the vehicle's current operating state. When the vehicle speed deviation is large or continues to increase, the controller improves its response capability and accelerates vehicle speed recovery; when the vehicle speed deviation is small or the error changes rapidly, the controller reduces the control intensity to suppress torque fluctuations and longitudinal impacts, thereby balancing response speed, vehicle speed stability, and comfort.

[0060] Sixth, this invention limits the original cruise torque request based on the motor's current maximum allowable drive torque and maximum energy recovery torque, ensuring that the output torque always remains within the motor's executable range. When the original cruise torque request exceeds the motor's maximum drive torque, it is limited to the maximum drive torque; when the original cruise torque request is less than the maximum energy recovery torque, it is limited to the maximum energy recovery torque, thereby ensuring that the constant speed cruise torque request matches the motor's actual execution capability.

[0061] Seventh, this invention uses the actual execution torque fed back from the power source as the input to the extended state observer, instead of using the original cruise torque request before the limit or the theoretical output of the controller as the observer input. This avoids the problem of inconsistency between the observer input and the actual force state of the vehicle caused by motor torque limiting, battery power limiting, energy recovery capability limiting, or vehicle safety strategy torque limiting, thus improving the accuracy of total disturbance estimation and cruise control stability.

[0062] Eighth, the present invention limits the rate of torque change before the cruise torque request output, and can use a smaller limit value for the rate of torque change in the zero-crossing region when switching between negative and positive torque, so that the process of releasing the recovered torque and building up the drive torque is smoother, reducing the impact on the power transmission system and improving the longitudinal smoothness of the vehicle when entering the cruise control process.

[0063] In summary, this invention can effectively improve problems such as discontinuous initial state of the controller, uneven torque connection, negative torque drag, vehicle speed fluctuation, and insufficient anti-disturbance capability during the vehicle cruise control switching process. Moreover, the required signals are all commonly used signals of the vehicle controller and can be implemented through the vehicle controller software, which has good engineering applicability and industrial application value.

[0064] 2. Other beneficial effects or advantages of the present invention will be described in detail in the specific embodiments. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] in:

[0067] Figure 1 This is a flowchart illustrating the steps of a torque initialization and adaptive active disturbance rejection control method for cruise control switching, provided by an exemplary embodiment of the present invention.

[0068] Figure 2This is a flowchart of a torque initialization and adaptive active disturbance rejection control method for cruise control switching provided by an exemplary embodiment of the present invention;

[0069] Figure 3 This is a schematic diagram of the AFLADRC cruise speed tracking simulation results provided by an exemplary embodiment of the present invention, wherein... Figure 3 The horizontal axis represents time in seconds, and the vertical axis represents vehicle speed in km / h.

[0070] Figure 4 This is a graph showing the actual vehicle's cruise control status and speed changes. The solid blue line represents the actual vehicle speed, the dashed blue line represents the set speed, and the orange curve indicates cruise control activation. Detailed Implementation

[0071] With the development of new energy vehicles, hybrid vehicles, and intelligent driving assistance technologies, cruise control has become a common function in vehicle longitudinal control systems. For pure electric vehicles or hybrid vehicles, cruise control is typically performed by the vehicle controller based on information such as the driver's target speed, the vehicle's actual speed, motor speed, motor feedback torque, gear status, brake pedal status, accelerator pedal status, battery power limit, motor torque capability, and vehicle fault status. The controller then sends the target torque request to the motor controller for execution, enabling the vehicle to travel stably around the set speed.

[0072] In existing vehicle cruise control methods, a common technical solution is to use PI or PID closed-loop control based on vehicle speed deviation. This type of method typically uses the deviation between the target vehicle speed and the actual vehicle speed as the control input. It calculates the basic cruise control quantity through proportional, integral, and derivative calculations, and then combines this with vehicle driving resistance, gradient compensation, acceleration compensation, torque feedforward parameters, or empirically calibrated parameters to obtain the target torque request from the motor. To adapt to different vehicle speed deviations or different acceleration states, some existing technologies also set multiple sets of PI / PID parameters, cruise demand torque tables, or segmented control logic, switching different control parameters or torque value methods when the vehicle speed deviation is in different ranges.

[0073] Furthermore, to improve torque fluctuations during cruise control, existing technologies typically add torque limiting, torque change rate limiting, low-pass filtering, or ramp transition processing to the controller output, ensuring that the torque request ultimately sent to the motor controller changes gradually according to a preset slope. For the control of new energy vehicles during downhill, coasting, or deceleration cruise conditions, some existing solutions also calculate the target deceleration torque or select an appropriate deceleration control method based on factors such as the regenerative braking capability, braking system capability, and the difference between the current vehicle speed and the target vehicle speed, in order to improve the stability of the cruise deceleration process.

[0074] The aforementioned existing technologies are effective in steady-state cruise or general speed tracking scenarios, but the following defects still exist at the moment when the vehicle switches from normal driving mode, coasting energy recovery mode or other longitudinal control modes to cruise control mode.

[0075] First, existing PI / PID cruise control methods typically focus primarily on closed-loop speed regulation after entering cruise control, with less consideration given to the continuity between the vehicle's actual torque and the cruise controller's initial output torque at the moment before entering cruise control. When the vehicle is in a driving, zero-torque coasting, or negative torque energy recovery state before entering cruise control, if the cruise controller directly uses zero torque, the calibrated torque, or the recalculated closed-loop torque as its initial output, it can easily lead to a sudden torque change at the moment of entering cruise control, causing longitudinal jerking, speed fluctuations, or driving discomfort.

[0076] Second, traditional PI / PID control is highly dependent on the calibration of control parameters. During actual vehicle operation, the vehicle's weight, load, road gradient, rolling resistance, wind resistance, battery power status, and motor output capability all change. Fixed-parameter PI / PID or segmented calibration tables cannot simultaneously balance response speed and comfort: when parameters are too large, the vehicle speed response is faster, but torque fluctuations, overshoot, or longitudinal shocks are likely to occur; when parameters are too small, the vehicle ride comfort is improved, but when there are significant changes in gradient, load, or external disturbances, problems such as slow speed recovery, increased steady-state error, or cruise speed fluctuations are likely to occur.

[0077] Third, existing output torque smoothing, slope limiting, or filtering methods are essentially post-processing measures. They can only limit the variation in the final torque request and cannot address the issue of discontinuous initial states at the moment of switching from the controller's internal state. Especially when cruise control is first engaged, if the controller's integral state, observer state, or disturbance estimation state is inconsistent with the vehicle's current true longitudinal dynamic state, even with slope limiting of the output torque, issues such as controller response lag, unreasonable compensation direction, or repeated torque corrections may still occur.

[0078] Fourth, in the actual control process of new energy vehicles, the original torque request calculated by the vehicle controller usually needs to be processed by limiting the maximum drive torque of the motor, the maximum regenerative torque, the battery charging and discharging power, thermal management, vehicle safety strategies, and the torque change rate. Therefore, the original torque request calculated by the controller is not necessarily equal to the torque actually executed by the power system. If the control algorithm or observer still uses the original torque request before limiting as input, it may cause inconsistency between the estimated state inside the controller and the actual force state of the vehicle, thereby reducing the vehicle speed tracking stability and disturbance rejection capability.

[0079] Fifth, while some existing robust control, adaptive control, or active disturbance rejection control methods can improve the vehicle speed tracking process's adaptability to model uncertainties and external disturbances, they typically focus more on vehicle speed tracking performance during continuous cruising and do not adequately consider the transient process of switching from other driving modes to cruise control. In particular, existing technologies have not yet provided targeted solutions for issues such as torque initialization upon entering cruise mode, initial state setting of the extended state observer, and consistency between the actual executed torque and the observer input after amplitude limiting.

[0080] Therefore, existing vehicle cruise control technology still cannot fully solve the problems that occur in new energy vehicles during the cruise control switching process, such as unreasonable initial torque, uneven torque connection, vehicle speed fluctuation, insufficient anti-disturbance capability, and inconsistency between the internal state of the controller and the actual executed torque. There is an urgent need for a cruise control switching method that can combine the actual executed torque before entering cruise, controller state initialization, and adaptive anti-disturbance control to improve the smoothness, stability, and engineering applicability of the cruise control entry process.

[0081] In view of this, the present invention provides a novel solution: a torque initialization and adaptive disturbance rejection control method for cruise control switching. This method is applied to the vehicle controller of pure electric vehicles or hybrid electric vehicles to address longitudinal jerking, torque abrupt changes, and vehicle speed fluctuations caused when a vehicle switches from normal driving mode, coasting energy recovery mode, or other longitudinal control modes to cruise control mode. These issues arise from discontinuous controller initial states, improper inheritance of the actual torque before entering cruise, insufficient cruise baseline torque compensation, inconsistencies between torque limits and actual torque, and changes in external disturbances such as vehicle mass and road gradient.

[0082] The technical concept of this invention lies in the following: At the instant the vehicle enters cruise control mode, instead of directly initializing the cruise controller output to zero torque, and also instead of directly adding the actual torque executed at the moment before entering cruise to the final output torque, the actual torque executed by the power source at the moment before entering cruise is used to initialize the extended state observer of the fuzzy adaptive linear active disturbance rejection controller. This allows the controller to output an initial control torque that matches the actual longitudinal dynamic state of the vehicle before entering cruise in the first control cycle. Simultaneously, the vehicle controller queries a pre-calibrated cruise baseline compensation torque table based on the current actual vehicle speed to obtain the cruise baseline compensation torque corresponding to the current speed. This cruise baseline compensation torque is then added to the AFLADRC output torque to form the original cruise torque request.

[0083] Once the vehicle enters cruise control, the vehicle controller limits the initial cruise torque request based on the motor's maximum allowable drive torque and maximum energy recovery torque. It then updates the extended state observer with the actual executed torque fed back from the power source, ensuring the observer input aligns with the vehicle's actual longitudinal force state. Simultaneously, the vehicle controller uses a fuzzy controller to online correct the Linear Active Disturbance Rejection Controller (LADRC) parameters based on the vehicle speed error and its rate of change. This allows the controller to adaptively adjust the control strength according to the vehicle's current deviation, thereby improving the smoothness of cruise control switching, vehicle speed stability, and disturbance rejection capability.

[0084] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0085] like Figure 1 As shown, the present invention provides a torque initialization and adaptive active disturbance rejection control method for cruise control switching, comprising the following steps:

[0086] Obtain the actual torque of the power source at the moment before the vehicle enters cruise control mode;

[0087] Based on the current actual vehicle speed, consult the cruise basic compensation torque table to obtain the cruise basic compensation torque;

[0088] The extended state observer in the fuzzy adaptive linear active disturbance rejection controller is initialized using the actual torque of the power source at the moment before entering cruise.

[0089] The cruise base compensation torque is added to the torque output by the fuzzy adaptive linear active disturbance rejection controller to generate the original cruise torque request.

[0090] Based on the motor’s current maximum allowable drive torque and maximum energy recovery torque, the original cruise torque request is limited to obtain the limited cruise torque request.

[0091] The extended state observer is updated using the actual execution torque fed back from the power source.

[0092] Through the above technical solution, firstly, the method of the present invention can effectively eliminate the torque surge during cruise control switching, achieving a smooth torque transition. Specifically, existing PI / PID cruise control typically sets the controller's initial output to zero torque or a recalculated closed-loop torque at the moment of entering cruise mode, without considering the actual torque of the vehicle at the previous moment. This leads to a step change in torque during switching, causing longitudinal jerking of the vehicle. This implementation, however, by "obtaining the actual torque of the power source at the moment before the vehicle enters cruise control mode" and "using this actual torque to initialize the extended state observer," ensures that the initial output of the fuzzy adaptive linear active disturbance rejection controller in the first control cycle naturally includes the true power state information of the vehicle before entering cruise control. This avoids the controller starting from a zero or unreasonable state, thereby eliminating the torque surge during switching and achieving a smooth transition of drive / recovery torque.

[0093] Secondly, the method of this invention can provide a basic torque feedforward that matches the current vehicle speed, enhancing the adaptability of cruise conditions. Specifically, existing methods mostly rely on pure closed-loop feedback or fixed calibration tables. When the vehicle speed changes, the basic torque requirement (such as overcoming rolling resistance and air resistance) cannot be adaptively adjusted in real time, resulting in excessive torque at low speeds or insufficient torque at high speeds. This implementation method, however, obtains the cruise basic compensation torque by "looking up the cruise basic compensation torque table based on the current actual vehicle speed" and "adding" this torque to the output of the active disturbance rejection controller to generate the original cruise request. This uses the basic compensation torque as a feedforward term, which can provide the basic torque required to maintain cruise based on the real-time vehicle speed, reducing the burden on feedback control and enabling the controller to respond quickly at different vehicle speeds, thus improving the adaptability of cruise conditions.

[0094] Third, the method of this invention ensures that the internal state of the controller is consistent with the actual force state of the vehicle, thus improving control accuracy. Specifically, in traditional active disturbance rejection control or PID control, the observer or integrator typically uses the theoretical target torque output by the controller as input. When the actual torque executed by the motor is less than the target request due to power limitations, thermal protection, or other reasons, the observer may mistakenly assume that the vehicle has obtained the expected driving force, leading to disturbance estimation deviation and vehicle speed tracking error. In this embodiment, by "updating the extended state observer using the actual executed torque fed back from the power source," the observer's input is made consistent with the vehicle's true longitudinal force state. This avoids the problem of observer input distortion caused by torque limiting or safety strategies, improves the accuracy of total disturbance estimation, and thus enhances the stability and anti-interference capability of cruise control.

[0095] Fourth, the method of this invention ensures that the cruise torque request is always within the motor's executable range, avoiding actuator saturation. Specifically, existing methods may output torque requests exceeding the motor's current maximum drive capacity or maximum regenerative braking capacity, leading to actuator saturation, preventing the actual vehicle speed from following the target speed, and even triggering system malfunctions. In this embodiment, the original cruise torque request is limited based on the motor's current maximum allowable drive torque and maximum energy recovery torque to obtain a limited cruise torque request. This ensures that the torque sent to the motor controller is always within the motor's real-time executable physical range, avoiding invalid torque requests and guaranteeing the feasibility of cruise control and system safety.

[0096] Fifth, the method of this invention comprehensively utilizes multi-source information to form a complete cruise switching and control closed loop. Specifically, existing related technologies often handle torque initialization, basic compensation, limiting, and observer updates in isolation, lacking coordination among the various stages and resulting in fragmented control logic during the switching process. In this embodiment, however, the steps of "acquiring actual torque before switching," "vehicle speed lookup table compensation," "observer state initialization," "torque superposition generation," "execution capability limiting," and "actual torque feedback updating the observer" are organically combined to form a complete control loop from state initialization, feedforward compensation, closed-loop correction, output limiting to observer feedback updating. Thus, through this closed-loop structure, the continuity of torque, the rationality of torque requests, and the authenticity of the observer state can be ensured during cruise switching, thereby effectively improving the smoothness, stability, and engineering reliability of the vehicle when entering cruise mode.

[0097] The method of the present invention will be further described below with reference to an exemplary embodiment.

[0098] Please see Figure 2 In this exemplary embodiment, the method of the present invention may include the following steps:

[0099] 1. Determine if the vehicle meets the conditions for entering cruise control.

[0100] The vehicle controller determines whether the vehicle meets the cruise control entry conditions based on the vehicle's operating status, driver's operation status, and powertrain status. When the vehicle meets the preset entry conditions, it is allowed to switch from the current longitudinal control mode to cruise control mode.

[0101] For example, cruise control activation conditions may include:

[0102] The cruise control function switch is active, and a valid trigger signal is detected for at least one of the cruise control setting switch, acceleration switch, or deceleration switch.

[0103] The current vehicle speed is within the preset allowable cruise speed range;

[0104] The current gear is drive;

[0105] The brake pedal was not depressed, and the vehicle was not in a state of active braking by the driver.

[0106] The accelerator pedal is not depressed, or the accelerator pedal opening is less than the preset control threshold.

[0107] The cruise control off switch was not triggered, and the cruise control exit signal was invalid.

[0108] The parking brake is released.

[0109] No faults affecting cruise control were detected in the motor system, battery system, and vehicle control system.

[0110] The vehicle's current driving mode allows cruise control to take over longitudinal control.

[0111] When all the above conditions are met, the vehicle controller sets the cruise control entry flag and enters the subsequent torque initialization and fuzzy adaptive anti-disturbance control process.

[0112] 2. Obtain the actual torque of the power source at the moment before entering cruise.

[0113] Before the vehicle enters cruise control mode, the vehicle controller obtains the actual torque feedback signal from the power source in real time via CAN communication. The actual torque feedback signal from the power source can be the actual output torque of the motor, the actual output torque of the engine, or the equivalent drive torque calculated based on the vehicle's powertrain structure.

[0114] In the control cycle preceding the vehicle's switch from the current driving mode to cruise control mode, the vehicle controller records the actual torque executed by the power source and denotes it as:

[0115]

[0116] in, The torque value that the power system actually outputs and participates in the longitudinal control of the vehicle just before the vehicle enters cruise control. The actual torque fed back by the power source during the control cycle prior to entering cruise; The control cycle for the vehicle to enter cruise control mode.

[0117] Understandable, It can be positive torque, zero torque, or negative torque. When the vehicle is in drive mode before entering cruise control... It is usually positive torque; when the vehicle is in a zero-torque coasting state... It can approach zero; when the vehicle is in coasting energy recovery mode, It can be negative torque.

[0118] because The actual torque feedback signal from the power source can reflect the true longitudinal power state of the vehicle just before entering cruise control. Therefore, this invention uses it as the basis for calculating the initial torque of cruise control to improve the continuity between the initial output of the cruise controller and the original power state of the vehicle.

[0119] 3. Calculate the cruise basic compensation torque by referring to the table based on the current vehicle speed.

[0120] The moment the vehicle enters cruise control mode, the vehicle controller obtains the current actual vehicle speed and queries the pre-calibrated cruise basic compensation torque table based on the current actual vehicle speed to obtain the cruise basic compensation torque corresponding to the current vehicle speed.

[0121] The cruise basic compensation torque table can be a one-dimensional vehicle speed lookup table, or it can be extended to a multi-dimensional lookup table with inputs of vehicle speed, gear, driving mode, estimated vehicle mass, or estimated road gradient, depending on the vehicle control requirements. Its basic form can be expressed as:

[0122]

[0123] in, The cruise basic compensation torque is obtained by looking up the table based on the current vehicle speed the moment the vehicle enters cruise control. The actual vehicle speed at the moment cruise control is activated; This is a lookup function for the pre-calibrated vehicle speed-cruise-based compensation torque.

[0124] The cruise baseline compensation torque characterizes the basic torque requirement needed for a vehicle to maintain cruise at the current speed. For pure electric vehicles or hybrid vehicles, this baseline compensation torque can be used to compensate for vehicle rolling resistance, air resistance, transmission resistance, and the negative torque drag effect when the vehicle switches from coasting energy recovery mode to cruise control mode. The cruise baseline compensation torque is not a standalone final cruise torque output, but rather participates in the cruise torque request calculation as a feedforward compensation term along with the AFLADRC output torque.

[0125] 4. Initialize the state of the linearly extended state observer.

[0126] At the moment the vehicle enters cruise control mode, this invention does not simply add the actual torque of the power source executed at the moment before entering cruise to the cruise basic compensation torque as the final output torque. Instead, it uses the actual torque of the power source executed at the moment before entering cruise to initialize the extended state observer (LESO) of the fuzzy adaptive linear active disturbance rejection controller.

[0127] The vehicle speed observation state of the linear extended state observer is initialized to the actual vehicle speed at the moment before entering cruise:

[0128]

[0129] in, The initial values ​​for the vehicle speed observation state of the extended state observer; The actual vehicle speed at the moment cruise control is activated.

[0130] The total disturbance observation state of the linear extended state observer is initialized based on the actual torque of the power source at the moment before entering cruise:

[0131]

[0132] in, The initial values ​​of the total perturbation observation state for the extended state observer; This is the nominal longitudinal dynamics input gain estimate used when the vehicle enters cruise mode; The actual torque of the power source in the control cycle before entering cruise control.

[0133] Using the above initialization method, the actual torque executed at the moment before cruise is entered. It is used to determine the initial total disturbance state of the extended state observer, so that the feedback correction torque output by the AFLADRC controller in the first control cycle of entering cruise can include information about the actual dynamic state of the vehicle before entering cruise, avoiding the controller from starting from a zero state or an unreasonable state, thereby improving the torque continuity at the moment of cruise control switching.

[0134] 5. Calculate the vehicle speed error and the rate of change of vehicle speed error.

[0135] After the vehicle enters cruise control mode, the vehicle controller periodically calculates the speed error between the target speed and the actual speed:

[0136]

[0137] And calculate the rate of change of vehicle speed error:

[0138]

[0139] in, This represents the vehicle speed error during the current control cycle. This represents the vehicle speed error from the previous control cycle. This represents the rate of change of vehicle speed error. The target vehicle speed for the current control cycle; The actual vehicle speed during the current control cycle; To control the cycle.

[0140] Speed ​​error The rate of change of vehicle speed error reflects the degree to which the vehicle's current speed deviates from the target speed. It reflects the trend of vehicle speed deviation. Both are used together to determine whether the current cruise control should favor a rapid response or a smooth suppression.

[0141] 6. Correct LADRC control parameters online using a fuzzy controller.

[0142] This invention employs a fuzzy controller, based on vehicle speed error. and vehicle speed error change rate The parameters of the linear active disturbance rejection controller are corrected online.

[0143] The input to the fuzzy controller is:

[0144]

[0145]

[0146] The output of the fuzzy controller includes the disturbance compensation coefficient correction, the extended state observer bandwidth correction, and the controller bandwidth correction, which are expressed as follows: , , :

[0147]

[0148]

[0149]

[0150] in, , , These are the basic calibration values ​​for the disturbance compensation coefficient, the extended state observer bandwidth, and the controller bandwidth, respectively. , , These are the disturbance compensation coefficient, extended state observer bandwidth, and controller bandwidth after correction in the k-th control cycle, respectively.

[0151] It is understood that in this invention, the disturbance compensation coefficient correction, the extended state observer bandwidth correction, and the controller bandwidth correction can all be updated in real time. That is, the current value is the sum of the base value and the correction calculated in fuzzy order at the previous moment.

[0152] When the vehicle speed error is large and the direction of error change indicates that the vehicle is deviating more from the target speed, the fuzzy controller increases the controller bandwidth and disturbance compensation intensity to improve the vehicle speed recovery speed; when the vehicle speed error is small or the error change rate is large, the fuzzy controller reduces the controller bandwidth or reduces the compensation intensity to suppress torque mutation and longitudinal impact.

[0153] Therefore, the present invention can adaptively adjust the control intensity according to the vehicle speed deviation state, thereby improving the adaptability of cruise control when the vehicle mass, load, road slope, rolling resistance and wind resistance change.

[0154] 7. A tracking differentiator is used to generate a smooth target vehicle speed.

[0155] To avoid sudden torque changes caused by abrupt changes in the target vehicle speed directly affecting the controller, this invention employs a tracking differentiator to smooth the target vehicle speed during constant speed cruise.

[0156] The discrete form of the tracking differentiator is:

[0157]

[0158]

[0159]

[0160]

[0161] in, For the first The cruise control set speed for each control cycle; To smoothly set the vehicle speed to track the output of the differentiator; To smooth out the rate of change of vehicle speed; To control the cycle; To track the bandwidth parameters of the differentiator; This is to track the smooth acceleration term calculated internally by the differentiator.

[0162] Through the above processing, the vehicle speed is smoothly set when the driver's set speed changes. It can gradually track the set vehicle speed according to the preset dynamic process. This avoids sudden changes in the set vehicle speed directly affecting subsequent controllers. The tracking differentiator in this invention is only used to smooth the set vehicle speed and generate a smoothed set vehicle speed and its rate of change, without using the acceleration signal fed back from the linear extended state observer to participate in the tracking differentiator calculation.

[0163] 8. Calculate linear state error feedback based on the smooth setting of vehicle speed.

[0164] After obtaining the smooth set speed, the vehicle controller calculates the vehicle speed tracking error based on the smooth set speed and the vehicle speed observation value output by the extended state observer:

[0165]

[0166] in, For the first Speed ​​tracking error per control cycle; To smoothly set the vehicle speed to track the output of the differentiator; The vehicle speed observation value is output by the linearly extended state observer.

[0167] The linear state error feedback controller calculates the basic control quantity based on the vehicle speed tracking error:

[0168]

[0169] in, This is the basic control quantity output by the linear state error feedback controller; This represents the controller gain for the current control cycle. The controller bandwidth can be corrected based on the fuzzy controller. The calculation yielded:

[0170]

[0171] In this way, the controller input no longer directly uses the step-change set speed, but uses the set speed smoothed by the tracking differentiator, thereby reducing the torque impact when the cruise set speed changes and improving the smoothness of vehicle longitudinal control.

[0172] 9. Use a linear extended state observer to estimate the total disturbance.

[0173] The linear extended state observer is used to estimate the actual vehicle speed and the total disturbance experienced during the vehicle's longitudinal motion. The total disturbance includes the combined effects of factors such as changes in vehicle mass, road gradient, rolling resistance, air resistance, transmission system resistance, and powertrain output capacity on the vehicle's longitudinal motion.

[0174] The discrete update form of the linearly extended state observer is:

[0175]

[0176]

[0177]

[0178] in, For the first Vehicle speed observation values ​​for each control cycle; For the first Total disturbance observations for each control cycle; For the first The actual vehicle speed during each control cycle; This is the observation error; The actual torque of the power source input to the linearly extended state observer; For the first The disturbance compensation coefficient after correction for each control cycle; For the observer gain of state estimation error feedback, For the observer gain of the extended perturbation state error feedback; To control the cycle.

[0179] The observer gain can be determined by the observer bandwidth:

[0180]

[0181]

[0182] By estimating the total vehicle disturbance using a linear extended state observer, this invention can dynamically compensate for changes in gradient, load, and driving resistance without relying on an accurate vehicle dynamics model, thereby improving the disturbance resistance of cruise control.

[0183] 10. Calculate the AFLADRC output torque based on the total disturbance compensation.

[0184] Based on the basic control quantity output by the linear state error feedback controller and the total disturbance estimated by the linear extended state observer, the output torque of the fuzzy adaptive linear active disturbance rejection controller is calculated:

[0185]

[0186] in, For the first The output torque of the fuzzy adaptive linear active disturbance rejection controller in each control cycle; This is the basic control quantity output by the linear state error feedback controller; The total perturbation estimated by the extended state observer.

[0187] In the first control cycle after the vehicle enters cruise control, since the extended state observer has already been initialized based on the vehicle speed and the actual torque executed by the power source at the moment before entering cruise control, the initial output torque of AFLADRC can be expressed as:

[0188]

[0189] Combination:

[0190]

[0191] It can be seen that the actual torque executed just before entering cruise control The initial total disturbance state has been incorporated into the AFLADRC initial output calculation via the extended state observer. Therefore, AFLADRC will not start from a zero torque state upon entering cruise control, but will instead generate the corresponding initial control torque based on the vehicle's actual dynamic state before entering cruise control.

[0192] When the actual vehicle speed is lower than the set speed, AFLADRC can output positive correction torque to increase the vehicle's driving force; when the actual vehicle speed is higher than the set speed or the vehicle is on a downhill or speeding condition, AFLADRC can output negative correction torque to reduce the driving torque or request energy recovery torque.

[0193] Through the above method, the present invention does not involve the actual torque executed before entering cruise. Instead of directly adding it to the final cruise torque request, it utilizes... Initialize the extended state observer so that the initial output of AFLADRC naturally includes information about the power state before entering cruise, thereby improving control continuity during cruise control switching.

[0194] 11. Generate the original cruise torque request by superimposing the cruise base compensation torque.

[0195] After the vehicle enters cruise control mode, the vehicle controller adds the cruise base compensation torque corresponding to the current vehicle speed to the AFLADRC output torque to generate the original cruise torque request without amplitude limiting processing.

[0196] In the first control cycle after the vehicle enters cruise control, the initial cruise torque request is expressed as:

[0197]

[0198] in, The initial cruise torque request without amplitude limiting during the first control cycle of entering cruise control; The cruise baseline compensation torque is obtained by looking up the odometer based on the current vehicle speed at the moment of entering cruise control. The initial output torque is calculated by AFLADRC based on the state of the expanded state observer after initialization.

[0199] In subsequent control cycles, the vehicle controller continues to add the cruise base compensation torque corresponding to the current vehicle speed to the AFLADRC output torque to generate the original cruise torque request for the current control cycle without amplitude limiting:

[0200]

[0201] in, For the first The original cruise torque request without amplitude limiting processing for each control cycle; This is the cruise base compensation torque corresponding to the current vehicle speed; Output torque for AFLADRC.

[0202] In this manner, the cruise baseline compensation torque serves as a feedforward compensation term, providing the baseline torque required to maintain cruise at the current vehicle speed; the AFLADRC output torque serves as a feedback correction term, used to correct the cruise torque in real time based on the vehicle speed error, the rate of change of error, and the total disturbance estimated by the extended state observer. The actual executed torque at the moment before entering cruise control... Instead of being directly added to the cruise baseline compensation torque, it participates in the AFLADRC initial output calculation through the extended state observer initialization, thereby achieving torque continuity at the moment of cruise switching.

[0203] 12. Limiting is performed based on the maximum driving torque and maximum energy recovery torque of the motor.

[0204] The vehicle controller limits the original cruise torque request that has not been limited, based on the maximum allowable drive torque and maximum energy recovery torque of the motor.

[0205] The limiting process can be represented as:

[0206]

[0207] in, For the cruise torque request after amplitude limiting; The original cruise torque request without amplitude limiting processing; This represents the maximum allowable drive torque of the motor in the current control cycle, and its value is positive. This is the maximum energy recovery torque allowed by the motor in the current control cycle, and its value is negative. This is the amplitude limiting function.

[0208]

[0209] The above limiting process ensures that the cruise torque request always remains within the range of the motor's currently executable torque. When the original cruise torque request is greater than the motor's maximum drive torque, it is limited to the motor's maximum drive torque; when the original cruise torque request is less than the motor's maximum energy recovery torque, it is limited to the motor's maximum energy recovery torque; when the original cruise torque request is between the maximum energy recovery torque and the maximum drive torque, the torque request remains unchanged.

[0210] Among them, the maximum driving torque of the motor and the maximum energy recovery torque of the motor can be determined in real time by the motor controller, battery management system or vehicle controller based on the motor capacity, motor speed, battery charging and discharging power limit, motor temperature, battery temperature, current vehicle speed and vehicle safety strategy.

[0211] Therefore, the cruise torque request generation method of the present invention can be summarized as follows: the extended state observer is initialized with the actual execution torque at the moment before entering cruise, so that AFLADRC generates an initial output torque with torque continuity at the moment of entering cruise; then the AFLADRC output torque is added to the cruise basic compensation torque obtained by looking up the table at the current vehicle speed to obtain the original cruise torque request; finally, the cruise torque request within the range that the motor can execute is obtained by limiting the maximum drive torque and the maximum energy recovery torque of the motor.

[0212] 13. Limit the rate of change of torque for cruise torque requests.

[0213] To further ensure smooth torque output, the vehicle controller limits the rate of change of the cruise torque request after it has been limited, thus obtaining the final torque request output to the power source controller.

[0214] The change in torque is expressed as:

[0215]

[0216] Limit the amount of torque variation:

[0217]

[0218] The final output torque is expressed as:

[0219]

[0220] in, The torque request is ultimately output to the power source controller; The output torque for the previous control cycle; This is the limit for the rate of change of torque increase; This is the limit for the rate of change of torque reduction.

[0221]

[0222] in, This is a preset zero-crossing torque threshold.

[0223] When the torque request is in the zero-crossing region, the vehicle controller uses a zero-crossing torque change rate limit that is lower than the normal torque change rate limit to limit the rate of change of the torque request.

[0224]

[0225] in, This is the limit for the rate of change of torque in the zero-crossing region.

[0226] By setting a smaller torque change rate limit in the zero-crossing region when switching from negative to positive torque, the shock to the powertrain during the release of regenerated torque and the establishment of drive torque can be reduced, resulting in a smoother change in longitudinal acceleration when the vehicle transitions from coasting energy recovery state to cruise control.

[0227] 14. Update the extended state observer using the actual executed torque.

[0228] Unlike traditional LADRC which directly uses the controller's raw output as the observer input, this invention, when updating the observer in the linear expansion state, does not use the original torque request before limiting, nor the target torque request after limiting. Instead, it uses the actual execution torque fed back from the power source as the observer input.

[0229]

[0230] in, The actual execution torque is input to the linearly extended state observer; This is the actual torque signal fed back from the power source controller.

[0231] By using the actual executed torque as the input to the extended state observer, the observer input can be kept consistent with the actual longitudinal force state of the vehicle, avoiding the observer input distortion caused by motor torque limiting, battery power limiting, regenerative torque limiting, or vehicle safety strategy torque limiting, thereby improving the accuracy of total disturbance estimation and the stability of cruise control.

[0232] When the vehicle controller calculates a high cruise torque request, but the actual torque delivered by the power source is less than the target torque request due to battery power limitations, motor thermal protection, or torque limiting by safety strategies, if the extended state observer still uses the target torque before the limit as input, the observer will mistakenly believe that the vehicle has obtained the corresponding driving force, thus causing a disturbance estimation error. This invention uses the actual torque delivered by the power source as the observer input, which avoids the above problem and ensures that the estimated state inside the controller is consistent with the actual driving state of the vehicle.

[0233] 15. Determine the conditions for cruise control to disengage.

[0234] When the vehicle continuously meets the cruise control conditions, the vehicle controller periodically performs the following processes: target speed smoothing, fuzzy adaptive parameter correction, AFLADRC output torque calculation, cruise basic compensation torque lookup, original cruise torque request generation, motor drive and regenerative torque limiting, torque change rate limiting, and extended state observer update.

[0235] The vehicle controller will disengage cruise control when any of the following conditions are detected:

[0236] The driver presses the brake pedal;

[0237] The driver presses the accelerator pedal deeply to take active control;

[0238] The cruise control off switch was triggered;

[0239] The current gear does not meet the cruise control requirements;

[0240] The vehicle's actual speed exceeds the permitted cruise speed range;

[0241] The motor system, battery system, or vehicle control system malfunctions that affect the cruise control function;

[0242] Other preset cruise exit conditions were triggered.

[0243] When exiting cruise control, the vehicle controller records the actual torque executed by the current power source, providing a basis for the initialization of the extended state observer when entering cruise control again.

[0244] The technical solution of the present invention will be further explained below with reference to an exemplary embodiment of the simulation verification process and the actual vehicle testing process.

[0245] I. Simulation verification.

[0246] In this simulation, it's important to note that the focus is not on re-describing the complete algorithm, but rather on combining simulation results with real-vehicle test results to illustrate the correspondence between the actual torque executed before entering cruise control, the current vehicle speed base compensation torque, the initialization of the AFLADRC extended state observer, and the final cruise torque output when the vehicle switches from coasting energy recovery state or low-speed deceleration state to cruise control. This demonstrates that the actual torque executed before entering cruise control is not simply added to the final output torque, but is used for the initialization of the extended state observer; the base compensation torque obtained from the current vehicle speed lookup table serves as a feedforward term in the generation of the original cruise torque; and the AFLADRC controller performs closed-loop correction of vehicle speed errors and external disturbances after entering cruise control, thus enabling the vehicle to smoothly transition from regenerative deceleration state to cruise control state.

[0247] During this simulation experiment, please refer to Figure 3The initial vehicle speed was approximately 16.3 km / h, and the target speed was then set to 20 km / h to simulate the speed recovery process after entering cruise control at low speeds. In the simulation curve, when the set speed increased from approximately 16.3 km / h to 20 km / h, the speed response output of the AFLADRC controller quickly followed the set speed and converged smoothly upon approaching the target speed, without significant oscillations or continuous overshoot.

[0248] The simulation results verify the feasibility of the AFLADRC vehicle speed closed-loop control process of this invention at the control algorithm level. In other words, when the target vehicle speed is higher than the current actual vehicle speed, AFLADRC can gradually establish cruise control parameters based on the vehicle speed error and its changing trend, smoothly bringing the actual vehicle speed closer to the set speed. These simulation results provide a foundation for algorithm-level verification in subsequent real-vehicle testing.

[0249] II. Real vehicle testing.

[0250] 1. Operating conditions description.

[0251] In real-world testing, cruise control was activated at an actual speed of approximately 16.3 km / h, with the set speed at approximately 20 km / h. Before activating cruise control, the vehicle was in a coasting energy recovery state, exhibiting negative torque drag and a deceleration tendency. At this time, the actual torque fed back by the motor before entering cruise control was -8 Nm, which reflects the vehicle's true longitudinal dynamic state before entering cruise control.

[0252] According to the method of the present invention, during the control cycle of entering cruise control, the vehicle controller first queries the cruise basic compensation torque table based on the current actual vehicle speed of 16.3 km / h to obtain the basic compensation torque corresponding to the current vehicle speed as 15 Nm. This basic compensation torque is used to characterize the basic drive requirements required for the vehicle to maintain cruise control near this speed, and to compensate for the negative torque drag effect when the vehicle switches from coasting energy recovery state to cruise control state.

[0253] 2. Verification process.

[0254] According to the control flow of the present invention, at the instant the vehicle enters cruise control, the vehicle speed observation state of the AFLADRC extended state observer is initialized to the current actual vehicle speed, approximately 16.3 km / h; simultaneously, the actual motor torque of -8 Nm at the moment before entering cruise control is used for the disturbance state initialization of the extended state observer. Through this initialization method, the initial state of the AFLADRC controller can inherit the true longitudinal dynamic state of the vehicle before entering cruise control.

[0255] Under this operating condition, the initial output of AFLADRC is approximately -8 Nm. This does not mean that -8 Nm is directly used as the final cruise output torque, nor does it mean that the negative torque before entering cruise is simply added to the final torque request. Instead, it means that through observer state initialization, the controller outputs a control quantity in the first control cycle that is continuous with the actual power state before entering cruise. This avoids the controller starting from a zero state or a state inconsistent with the vehicle's current force state, thereby reducing the sudden torque change during cruise switching.

[0256] Subsequently, the vehicle controller combines the 15Nm cruise baseline compensation torque obtained from the current vehicle speed lookup table with the AFLADRC output torque to form the initial cruise torque request. This initial cruise torque request is then processed by torque change rate limiting, motor drive torque upper limit, and energy recovery torque lower limit to obtain the final cruise torque request output to the motor controller. Thus, the actual execution torque before entering cruise mainly affects the initial output of AFLADRC through the initialization of the extended state observer, while the current vehicle speed baseline compensation torque participates in cruise torque generation as a feedforward compensation term. Together, they complete the smooth torque transition during the cruise entry process.

[0257] 3. Explanation of actual vehicle verification results.

[0258] Please see Figure 4 As can be seen from the actual vehicle curve, before cruise control was activated, the vehicle's actual speed was decreasing, indicating that the vehicle was in a state of coasting energy recovery or deceleration before entering cruise control. After cruise control was activated, the vehicle's actual speed did not suddenly increase or decrease, but gradually increased under the combined effect of torque change rate limitation and vehicle inertia. After about 7.3 seconds, the actual speed reached around 19 km / h, with an error of about 1 km / h compared to the set speed of about 20 km / h.

[0259] The results demonstrate that, under conditions where the actual torque is negative before entering cruise control and the current vehicle speed is lower than the set speed, the method of this invention can smoothly transition the vehicle from a regenerative deceleration state to a cruise control acceleration recovery state through a process of "initializing the actual torque with an expanded state observer, compensating based on the current vehicle speed lookup table, applying AFLADRC closed-loop correction, and limiting and restricting the rate of torque change." This process verifies the engineering feasibility of the formula flow and control logic of this invention, and also demonstrates the effectiveness of the method in reducing torque spikes during cruise control switching, improving the smoothness of speed recovery, and enhancing driving comfort.

[0260] 4. Conclusion.

[0261] During real-vehicle verification, the vehicle transitioned from coasting energy recovery mode to cruise control at an actual speed of approximately 16.3 km / h, with the set speed at approximately 20 km / h. In the control cycle prior to entering cruise control, the actual torque fed back by the motor was -8 Nm. The vehicle controller consulted the cruise baseline compensation torque table based on the current actual speed and obtained a baseline compensation torque of 15 Nm. Upon entering cruise control, the vehicle controller initialized the speed observation state of the extended state observer to the current actual speed and initialized the disturbance state using the actual torque of the motor at the moment prior to entering cruise control, resulting in an initial AFLADRC output of approximately -8 Nm. Subsequently, the AFLADRC output was combined with the 15 Nm baseline compensation torque to form the original cruise torque request, which was then output after torque change rate limiting and torque amplitude limiting processing. Simulation results show that the AFLADRC controller can smoothly track the vehicle speed from approximately 16.3 km / h to near the set speed of 20 km / h. Real-vehicle testing results show that after cruise control is activated, the actual vehicle speed increases smoothly, reaching approximately 19 km / h after about 7.3 seconds, close to the set speed. These results collectively verify that the method of this invention can reduce torque abrupt changes and improve the smoothness of speed recovery when switching from coasting energy recovery state to cruise control.

[0262] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A torque initialization and adaptive active disturbance rejection control method for cruise control switching, characterized in that, The process for switching a vehicle from coasting energy recovery mode to cruise control mode includes the following steps: The actual torque delivered by the power source at the moment before the vehicle enters cruise control mode is obtained and denoted as... ,in, This refers to the actual torque output of the powertrain that participates in the longitudinal control of the vehicle just before it enters cruise control. The actual torque fed back by the power source during the control cycle prior to entering cruise; Based on the current actual vehicle speed, consult the cruise basic compensation torque table to obtain the cruise basic compensation torque; Using the actual torque of the power source at the moment before entering cruise control, the extended state observer in the fuzzy adaptive linear active disturbance rejection controller is initialized, including: initializing the vehicle speed observation state of the extended state observer to the actual vehicle speed at the moment before entering cruise control. Initialize the total perturbation observation state of the extended state observer to ,in, This is the initial value for the vehicle speed observation state. The actual vehicle speed at the moment cruise control is activated. The initial values ​​for the total disturbance observation state, This is the nominal longitudinal dynamics input gain estimate used when the vehicle enters cruise control. The actual torque delivered by the power source at the moment before entering cruise control; The cruise baseline compensation torque is added to the torque output by the fuzzy adaptive linear active disturbance rejection controller to generate the original cruise torque request; wherein, in the first control cycle after the vehicle enters cruise control... The original cruise torque request is obtained by adding the cruise base compensation torque at the moment of entering cruise and the initial output torque calculated based on the state of the expanded state observer after initialization. Based on the current maximum allowable driving torque and maximum energy recovery torque of the motor, the original cruise torque request is limited to obtain a limited cruise torque request. The actual execution torque fed back from the power source is used, and when updating the extended state observer, the actual execution torque fed back from the power source is used as the input to update the extended state observer.

2. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 1, characterized in that, The step of querying the cruise basic compensation torque table based on the current actual vehicle speed further includes: The vehicle's actual speed is obtained the instant it enters cruise control mode. and according to Query the pre-calibrated vehicle speed-cruise baseline compensation torque table to obtain the cruise baseline compensation torque corresponding to the current vehicle speed. , represented as: In the formula, This is a lookup function for the pre-calibrated vehicle speed-cruise baseline compensation torque.

3. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 1, characterized in that, It also includes the step of correcting the parameters of the linear active disturbance rejection controller online using a fuzzy controller: Based on vehicle speed error and vehicle speed error change rate The disturbance compensation coefficient correction amount is output through the fuzzy controller. , bandwidth correction of extended state observer and controller bandwidth correction And calculate the corrected parameters: In the formula, , and These are the basic calibration values ​​for the disturbance compensation coefficient, the extended state observer bandwidth, and the controller bandwidth, respectively. , , These represent the disturbance compensation coefficient corrected in the k-th control cycle, the extended state observer bandwidth, and the controller bandwidth, respectively, and the vehicle speed error. Target speed With actual vehicle speed The difference, the rate of change of the vehicle speed error The difference between the current cycle speed error and the previous cycle speed error is used to determine the control cycle. The ratio is calculated.

4. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 3, characterized in that, Before calculating the vehicle speed error, the process also includes a step of smoothing the target vehicle speed for cruise control using a tracking differentiator: The target vehicle speed is obtained by using a discrete form of a tracking differentiator. Processing is performed to obtain a smooth set speed. and its rate of change Specifically: In the formula, For the first The cruise control sets the vehicle speed for each control cycle. To smoothly set the vehicle speed to track the output of the differentiator, To smoothly set the rate of change of vehicle speed, To control the cycle, To track the differentiator bandwidth parameter, This is to track the smooth acceleration term calculated internally by the differentiator.

5. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 4, characterized in that, The step of outputting torque by the fuzzy adaptive linear active disturbance rejection controller further includes: Based on smooth speed setting Vehicle speed observations output by the extended state observer Calculate vehicle speed tracking error ; The basic control quantity is calculated based on the vehicle speed tracking error. , where controller gain By the corrected controller bandwidth Determined, that is ; The total disturbance estimated based on the basic control quantity and the extended state observer. Calculate the output torque of the fuzzy adaptive linear active disturbance rejection controller. : In the formula, This is the corrected disturbance compensation coefficient.

6. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 1, characterized in that, The step of generating the original cruise torque request further includes: The cruise base compensation torque corresponding to the current vehicle speed With fuzzy adaptive linear active disturbance rejection controller output torque Add them together to get the original cruise torque request without the limiting process. : .

7. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 1, characterized in that, The step of limiting the original cruise torque request further includes: The limiting process is represented as follows: In the formula, The limiting function is defined as follows: ;in, For the cruise torque request after limiting, The original cruise torque request without limiting processing. This represents the maximum allowable drive torque of the motor in the current control cycle, and its value must be positive. This is the maximum energy recovery torque allowed by the motor in the current control cycle, and its value is negative.

8. The torque initialization and adaptive active disturbance rejection control method for cruise control switching according to claim 1, characterized in that, In the step of updating the extended state observer using the actual execution torque feedback from the power source, the discrete update form of the linear extended state observer is as follows: ;in, For observation error, For the first Vehicle speed observation values ​​for each control cycle For the first The actual vehicle speed during each control cycle To control the cycle, For the first Total disturbance observations for each control cycle For the first The disturbance compensation coefficient after each control cycle correction. The actual torque of the power source input to the linearly extended state observer, and , The actual torque signal fed back by the power source controller. For the observer gain of state estimation error feedback, The observer gain is used for feedback of the extended perturbation state error.

Images