Commercial vehicle master cylinder cooperative brake control system and method

Abstract

This application provides a tractor-trailer coordinated braking control system and method for commercial vehicles, belonging to the field of commercial vehicle braking control technology. The system includes: acquiring the vehicle's overall state and total demand; a central collaborative decision-maker decomposing tasks based on the total demand and allocating tasks to various agents; conducting capability assessment and bidding based on the vehicle's state; determining the output contracts of each agent according to the bidding operation contract network protocol; constructing a disturbance model library; estimating various disturbance sources based on a virtual disturbance observer; identifying major disturbance components; generating suppression signals; correcting the output contracts; and obtaining the final contract instruction; the central decision-maker generating a predicted trajectory for the tractor based on the final contract instruction and sending it to the trailer braking agent for trailer feedforward control and model reference adaptive feedback control, outputting trailer braking commands, and controlling the trailer's EBS execution. This application overcomes the challenge of composite jitter suppression, improving driving stability, comfort, and system robustness.

Description

Technical Field

[0001] This application relates to the field of commercial vehicle braking control technology, and in particular to a driver-trailer coordinated braking control system and method for commercial vehicles. Background Technology

[0002] Modern commercial vehicle braking systems have evolved from single friction braking to multi-source composite systems. Existing technologies regarding ride comfort control have the following main limitations:

[0003] (1) Inherent defects of traditional collaborative control architecture: Existing commercial vehicle braking control systems mostly adopt a "centralized command-decentralized execution" architecture. The central controller calculates the target output of each brake actuator based on the overall vehicle status and issues commands through the vehicle network. This architecture has an inherent flaw: the central controller struggles to accurately grasp the dynamic capability boundaries of each actuator in real time (such as the real-time pressure build-up capability of the EBS, the thermal state of the retarder, and the instantaneous torque fluctuation characteristics). This can lead to issued commands that exceed the actuator's current actual response capability, resulting in command lag, overshoot, or oscillation. This information asymmetry is the theoretical root cause of poor smoothness in multi-actuator systems.

[0004] (2) Shortcomings of multi-source disturbance suppression technology: To address the vibration issue during braking, existing technologies mostly employ local compensation strategies. For example, they suppress inherent torque ripple by improving the retarder excitation current algorithm and compensate for the hysteresis nonlinearity of the EBS valve through feedforward compensation. However, these methods are performed independently and lack consideration for the combined effects of multi-source disturbances. Disturbances of different frequencies and phases, such as EBS pressure fluctuations (frequency band 10-50Hz), retarder electromagnetic harmonics (frequency band 1-10Hz), and transmission system torsional vibration (frequency band 2-8Hz), can generate new and complex low-frequency or beat frequency oscillations when combined at the vehicle level, which existing single-point suppression technologies are ineffective against.

[0005] (3) The theoretical gap in smooth coordination between the main trailer and the trailer: Regarding the coordination of braking between the tractor and trailer, current mainstream technologies are still based on simple pressure or deceleration following strategies. For example, the ISO 11992 standard defines a communication protocol for the tractor to send target pressure commands to the trailer. However, this "command-response" model has significant delays and does not consider the coupling effect of the complex control strategies adopted by the tractor to achieve its own ride comfort (such as smooth transfer of braking force) on the trailer dynamics. When the tractor's braking strategy is dynamically adjusted to optimize ride comfort, the trailer control system cannot predict its adjustment intentions and generates hysteresis tracking errors, resulting in unexpected longitudinal or lateral interaction forces between the tractor and trailer, thus compromising overall ride comfort.

[0006] Both academia and industry lack an effective method to enable trailers to anticipate and coordinate with the smooth braking intentions of the tractor. Summary of the Invention

[0007] In view of this, embodiments of this application provide a driver-trailer coordinated braking control system and method for commercial vehicles, with the aim of overcoming the above-mentioned defects in the smoothness control of existing commercial vehicle braking systems.

[0008] In a first aspect, embodiments of this application provide a method for controlling the coordinated braking of a commercial vehicle driver and trailer. The method is implemented based on a coordinated braking control system for commercial vehicles, which includes a central coordinated decision-maker, an EBS intelligent agent, a retarder intelligent agent, and a trailer braking intelligent agent. The method includes: Obtain the overall vehicle status and total requirements; The central collaborative decision-maker decomposes tasks according to the overall demand and assigns tasks to each intelligent agent. Each intelligent agent conducts capability assessment and bidding based on the overall vehicle status. The central collaborative decision-maker runs the contract network protocol according to the bidding content to determine the output contract of each intelligent agent. A disturbance model library is constructed, and each disturbance source is estimated based on a virtual disturbance observer. The main disturbance components are identified, and suppression signals are generated based on the main disturbance components. The output contracts of each agent are corrected based on the suppression signals to obtain the final contract instructions of each agent. The central decision-maker generates a predicted trajectory for the tractor unit based on the final contract instructions, and sends the predicted trajectory to the trailer braking agent. The trailer braking agent performs trailer feedforward control and model reference adaptive feedback control, outputs trailer braking commands, and controls the trailer EBS to execute.

[0009] According to a specific implementation of an embodiment of this application, the central collaborative decision-maker decomposes tasks based on total demand and assigns tasks to various intelligent agents. Each intelligent agent performs capability assessment and bidding based on the overall vehicle status, including: Based on the overall braking force requirements and vehicle ride comfort goals, the central decision-maker decomposes the braking task into multiple sub-tasks. Each sub-task has service quality requirements, and the central decision-maker broadcasts the announcements of each sub-task to each intelligent agent. The EBS agent receives announcements from each subtask, assesses the current cylinder pressure, valve body temperature, and available accessory power based on its current dynamic state, calculates the maximum or minimum pressure change curve that it can stably provide in the next decision cycle, and includes a confidence index. The retarder agent receives announcements from each subtask, assesses the current rotor temperature, speed and cooling conditions based on its current dynamic state, calculates the torque curve it provides smoothly over a future decision cycle, and includes the spectral characteristics of its torque fluctuations. The EBS agent and the retarder agent each bid on the computational content.

[0010] According to a specific implementation of an embodiment of this application, the central collaborative decision-maker operates the contract network protocol based on the bidding content to determine the output contracts of each intelligent agent, including: After receiving all bids, the central decision-maker runs a collaborative planning algorithm. The collaborative planning algorithm aims to minimize the overall impact and balance the load of each agent. Under the premise of meeting the service quality requirements of each sub-task, it generates an output contract for each winning agent. The output contract includes the agent's target output curve and corresponding reward and penalty clauses for a future decision cycle. The central decision-making unit and the winning intelligent agent signed a power contribution contract.

[0011] According to a specific implementation of an embodiment of this application, the step of constructing a disturbance model library and estimating each disturbance source based on a virtual disturbance observer includes: A parameterized disturbance model library is established for EBS, retarder, and transmission system as disturbance sources; The virtual disturbance observer uses the longitudinal acceleration sensor signal of the whole vehicle as the total disturbance output, and combines it with the current contractual instructions of each agent. Through a parallel adaptive Kalman filter, it estimates the amplitude, frequency and phase of each disturbance source in real time.

[0012] According to a specific implementation of an embodiment of this application, the step of identifying the main disturbance component and generating a suppression signal based on the main disturbance component includes: Real-time spectrum analysis is performed on the estimated amplitude, frequency and phase of each disturbance source to identify the main disturbance components, which include disturbance amplitude and disturbance frequency. The suppression decision module dynamically configures a set of multi-channel adaptive notch filters based on the perturbation frequency distribution. Each channel of the adaptive notch filter corresponds to a perturbation frequency. Each channel calculates the suppression signal that needs to be injected into the control command of the agent corresponding to that channel. The phase of the suppression signal is opposite to the estimated phase, and the amplitude of the suppression signal is adaptively adjusted according to the disturbance amplitude and the controller gain.

[0013] According to a specific implementation of an embodiment of this application, the step of modifying the output contract of each agent based on the suppression signal includes: The suppression signal is superimposed on the original contract instruction of the corresponding intelligent agent to modify the output contract.

[0014] According to a specific implementation of an embodiment of this application, the central decision-maker generates a predicted trajectory for the tractor unit based on the final contract instruction, including: Based on the current vehicle status and the final contract instructions of each agent, the central collaborative decision-maker uses the vehicle's longitudinal dynamics model to generate the tractor's predicted trajectory within a preset time period. The tractor's predicted trajectory includes the predicted longitudinal deceleration and the desired articulation angle.

[0015] According to a specific implementation of an embodiment of this application, the trailer braking intelligent agent performs trailer feedforward control and model reference adaptive feedback control, and outputs trailer braking commands, including: The trailer braking agent receives the predicted trajectory of the tractor and uses it as a reference input for the feedforward controller. Using the simplified dynamic model of the trailer itself, the feedforward controller outputs the basic braking force of the trailer required to track the predicted trajectory of the tractor. A model reference adaptive controller is constructed. The reference model of the model reference adaptive controller is the predicted trajectory of the tractor, the feedback is the actual motion state of the trailer, the adaptive law is to minimize the tracking error and identify and compensate the nonlinear parameters at the connection between the tractor and the trailer online, and the output is the feedback correction force. The trailer braking intelligent agent combines the feedback correction force with the trailer's basic braking force to obtain the trailer braking command.

[0016] According to a specific implementation of the embodiments of this application, the actual motion state of the trailer includes the measured longitudinal deceleration of the trailer and the measured articulation angle, and the tracking error includes the error between the measured longitudinal deceleration of the trailer and the predicted longitudinal deceleration of the tractor and the error between the measured articulation angle and the expected articulation angle.

[0017] Secondly, this application also provides a commercial vehicle driver-trailer coordinated braking control system for executing the commercial vehicle driver-trailer coordinated braking control method as described in any embodiment of the first aspect. The system includes a central coordinated decision-maker, a braking execution intelligent agent cluster, a perception layer, and a network communication layer. The central coordinated decision-maker is used to run a multi-agent cooperative optimization algorithm and output trailer braking commands. The perception layer is used to collect vehicle status information, road environment information, and driver intention input information. The network communication layer is used to provide a communication network. The braking execution intelligent agent cluster includes an EBS intelligent agent, a retarder intelligent agent, and a trailer brake intelligent agent. The EBS intelligent agent is located in the EBS controller and is used for local pressure closed-loop control, as well as for evaluating and reporting its own real-time status and capability boundaries. The retarder intelligent agent is located in the retarder controller and is used for local torque control, as well as for evaluating and reporting thermal status and torque fluctuation characteristics. The trailer brake intelligent agent is located in the trailer EBS gateway or controller and is used to receive trailer braking commands and control trailer braking.

[0018] Beneficial effects: The commercial vehicle driver-trailer coordinated braking control system and method in this application embodiment have the following beneficial effects: Architectural breakthroughs enhance system robustness: Through a multi-agent negotiation mechanism, dynamic and flexible allocation of braking tasks is achieved, enabling the system to adapt to changes in actuator state, fundamentally avoiding conflicts between instructions and capabilities, and significantly improving the smoothness and robustness of control under complex working conditions. Overcoming the challenge of composite vibration suppression: The proposed virtual sensing and multi-channel collaborative suppression method can effectively address the problem of disturbance synthesis from different physical sources and different frequency bands, achieving a vehicle-level "ultra-smooth" braking feel that traditional methods cannot achieve, especially improving the smoothness on low-adhesion roads or during aggressive driving. Achieving forward-looking integrated control of master and trailer: Based on the prediction-based trailer coordination method, the trailer changes from a passive and lagging response to an active and synchronous cooperation, which basically eliminates the longitudinal impulse and lateral sway of the master and trailer caused by coordination delay, and improves the driving stability and comfort of the entire vehicle convoy. A complete technical closed loop is formed: by organically combining the three key links of negotiation allocation, disturbance suppression and motion coordination, a comprehensive solution for smooth braking with complete theory, logical consistency and engineering implementation is formed, which has significant practical value. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a diagram illustrating the architecture of a commercial vehicle driver-trailer coordinated braking control system according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the task negotiation and allocation process based on the Dynamic Contract Network protocol according to an embodiment of the present invention. Figure 3 This is a block diagram illustrating the principle of a multi-source synthetic jitter virtual observation and cooperative suppression system according to an embodiment of the present invention. Figure 4 This is a block diagram of a trailer feedforward-model reference adaptive cooperative control based on tractor motion prediction according to an embodiment of the present invention. Detailed Implementation

[0021] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0022] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this application, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0024] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. The illustrations only show the components related to this application and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0025] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0026] The applicant's research found that existing technologies either focus on optimizing the performance of a single actuator or employ simple rule / centralized methods to handle multi-system coordination. Neither provides a complete solution from the perspectives of system architecture and control theory that can simultaneously address the three core smoothness challenges: "dynamic task allocation under information asymmetry," "multi-band synthetic disturbance suppression," and "proactive motion coordination between the main unit and trailer." This invention aims to fill this technological gap.

[0027] In a first aspect, embodiments of this application provide a method for controlling the coordinated braking of a commercial vehicle driver and trailer. The method is implemented based on a coordinated braking control system for commercial vehicles, which includes a central coordinated decision-maker, an EBS intelligent agent, a retarder intelligent agent, and a trailer braking intelligent agent. The method includes: Step 1: Obtain the overall vehicle status and total requirements; Step 2: The central collaborative decision-maker decomposes the tasks according to the overall requirements and assigns the tasks to each intelligent agent. Each intelligent agent conducts capability assessment and bidding based on the overall vehicle status. The central collaborative decision-maker runs the contract network protocol according to the bidding content to determine the output contract of each intelligent agent. Step 3: Construct a disturbance model library, estimate each disturbance source based on a virtual disturbance observer, identify the main disturbance components, generate suppression signals based on the main disturbance components, and modify the output contracts of each agent based on the suppression signals to obtain the final contract instructions of each agent. Step 4: The central decision-maker generates the tractor prediction trajectory based on the final contract instructions and sends the tractor prediction trajectory to the trailer braking agent. The trailer braking agent performs trailer feedforward control and model reference adaptive feedback control, outputs trailer braking commands, and controls the trailer EBS to execute.

[0028] In practice, during each control cycle (e.g., 10ms), the system executes the above steps sequentially, all instructions are issued synchronously, each agent executes them, and the system enters the next cycle.

[0029] In this embodiment, a novel braking system control architecture based on multi-agent negotiation is constructed to achieve dynamic capability matching and flexible task allocation between the central decision-making unit and each braking execution unit, fundamentally eliminating the smoothness degradation caused by command-capability mismatch. A multi-source disturbance collaborative observation and suppression mechanism is proposed, which can identify and collaboratively cancel composite disturbances from different braking actuators online, solving the problem of synthetic jitter that is ineffective with traditional single suppression methods. A trailer feedforward-adaptive collaborative control strategy based on tractor motion intention prediction is designed, enabling the trailer braking system to coordinate with the tractor's smooth braking action in advance and accurately, achieving a smooth braking experience for articulated trains. Finally, a complete and implementable "negotiation-observation-prediction-collaboration" smooth braking control solution is formed, significantly improving the driving comfort, cargo safety, and driving stability of commercial vehicles.

[0030] The following section provides a detailed description of the multi-source task negotiation and allocation strategy based on the dynamic contract network protocol developed in step 2.

[0031] In one embodiment, the central collaborative decision-maker decomposes tasks according to total demand and assigns tasks to various intelligent agents. Each intelligent agent performs capability assessment and bidding based on the overall vehicle status, including: Based on the overall braking force requirements and vehicle ride comfort goals, the central decision-maker decomposes the braking task into multiple sub-tasks. Each sub-task has service quality requirements, and the central decision-maker broadcasts the announcements of each sub-task to each intelligent agent. The EBS agent receives announcements from each subtask, assesses the current cylinder pressure, valve body temperature, and available accessory power based on its current dynamic state, calculates the maximum or minimum pressure change curve that it can stably provide in the next decision cycle, and includes a confidence index. The retarder agent receives announcements from each subtask, assesses the current rotor temperature, speed and cooling conditions based on its current dynamic state, calculates the torque curve it provides smoothly over a future decision cycle, and includes the spectral characteristics of its torque fluctuations. The EBS agent and the retarder agent each bid on the computational content.

[0032] Furthermore, the central collaborative decision-maker operates the contract network protocol based on the bidding content to determine the output contracts of each intelligent agent, including: After receiving all bids, the central decision-maker runs a collaborative planning algorithm. The collaborative planning algorithm aims to minimize the overall impact and balance the load of each agent. Under the premise of meeting the service quality requirements of each sub-task, it generates an output contract for each winning agent. The output contract includes the agent's target output curve and corresponding reward and penalty clauses for a future decision cycle. The central decision-making unit and the winning intelligent agent signed a power contribution contract.

[0033] This embodiment details a multi-action source task negotiation and allocation strategy based on the Dynamic Contract Network protocol. This strategy reconstructs the relationship between the central decision-maker and each action execution agent into a multi-agent system conforming to the "manager-contractor" model. A specific embodiment is provided below to illustrate this strategy, with reference to... Figure 2 This includes the following steps: Task breakdown and announcement: The central decision-maker determines the overall braking force requirements. And vehicle ride comfort targets (such as maximum permissible impact) J max The braking task is broken down into multiple subtasks. Each subtask has accompanying quality of service requirements, such as "providing a smooth base braking force with a volatility of less than 5%" and "providing rapid dynamic compensation force with a response time of <100ms." Subsequently, the central decision-maker broadcasts the task announcement to all braking execution agents. Agent Capability Assessment and Bidding: After receiving the announcement, each agent will assess its capabilities and submit a bid based on its current dynamic state, including: EBS agent: assesses current cylinder pressure, valve body temperature, available accessory power, etc., calculates the maximum / minimum pressure change curve that it can stably provide in the next decision cycle, and includes a confidence index; The retarder agent assesses the current rotor temperature, speed, and cooling conditions, calculates the torque curve it can stably provide over a future decision cycle, and includes the spectral characteristics of its torque fluctuations. Contract Signing and Collaborative Planning: After receiving all bids, the central decision-maker runs a collaborative planning algorithm. This algorithm aims to minimize the overall impact and balance the load on each agent. While meeting the service quality requirements of each sub-task, it generates a power output contract for each winning agent. The contract clearly specifies the agent's target power output curve for a future period. And corresponding reward / penalty clauses (such as tracking accuracy rewards, exceeding limits penalties). The central decision-maker signs a contract with the winning intelligent agent; Contract execution and renegotiation: Each agent executes the contract. The central decision-maker continuously monitors the contract execution and the overall vehicle status. If a significant deviation occurs (such as a sudden drop in the capabilities of an agent) or a sudden change in task requirements, the contract renegotiation process is triggered to dynamically adjust task allocation.

[0034] The multi-braking source task negotiation and allocation strategy based on the dynamic contract network protocol in this embodiment effectively solves the problem of rigid binding between decision-making and execution in traditional centralized control by transforming the negotiation and allocation process of braking tasks into a "contractual agreement" model with clearly defined rights and responsibilities. This strategy grants each braking execution agent a certain degree of autonomous decision-making space, enabling it to participate in task bidding based on its own dynamic capabilities, while the central decision-maker achieves flexible control over the global collaborative process through the formulation and execution supervision of contract terms. This "decentralized decision-making, centralized coordination" mechanism not only improves the accuracy and dynamic adaptability of task allocation among multiple braking sources, but also provides a new theoretical framework for solving the smooth coordination problem of the main trailer under complex working conditions (such as sudden changes in road adhesion coefficient, partial failure of braking system, etc.), filling the theoretical gap in existing research that lacks a quantitative correlation between dynamic negotiation mechanisms and smoothness targets.

[0035] The following section provides a detailed description of the multi-source synthetic jitter collaborative suppression method based on virtual sensing and adaptive filtering developed in step 3, referring to... Figure 3 .

[0036] In one embodiment, constructing a disturbance model library and estimating each disturbance source based on a virtual disturbance observer includes: For EBS ( d ebs ), retarder ( d ret ) and transmission system ( ddrv Establish a parameterized disturbance model library as a disturbance source; When running online, the virtual disturbance observer utilizes signals from the vehicle's longitudinal acceleration sensor. As the total perturbation output, combined with the current contractual instructions of each agent... u i (As a known input), each disturbance source is estimated in real time by running a set of parallel adaptive Kalman filters. The amplitude, frequency, and phase.

[0037] In this embodiment, multi-source disturbance modeling and virtual sensing are implemented. This is equivalent to creating a "virtual sensor" for each potential physical disturbance source, enabling it to accurately capture the dynamic characteristics of each disturbance source without relying on direct physical sensor measurements. This disturbance estimation method based on a model library and adaptive filtering overcomes the limitations of traditional disturbance observation that relies solely on single sensor signals or simplified models. It achieves multi-source fusion online estimation of key disturbance sources such as EBS, retarder, and transmission system, providing a high-precision state awareness foundation for active disturbance compensation in subsequent main-trailer braking coordinated control.

[0038] Furthermore, the identification of the main disturbance components and the generation of a suppression signal based on the main disturbance components include: Real-time spectral analysis was performed on the estimated amplitude, frequency, and phase of each disturbance source to identify the main disturbance components, including the disturbance amplitude. A k and disturbance frequency f k ; The suppression decision module dynamically configures a set of multi-channel adaptive notch filters based on the perturbation frequency distribution. Each channel of the adaptive notch filter corresponds to a perturbation frequency. f k ; Each channel calculates the suppression signal that needs to be injected into the agent control command corresponding to that channel. The phase of the suppressed signal is opposite to the estimated phase, and the amplitude of the suppressed signal is based on the disturbance amplitude. A k And adaptive adjustment of controller gain.

[0039] In this embodiment, disturbance spectrum analysis and collaborative suppression decisions were performed. By decomposing multi-source disturbances into different frequency components and configuring notch filters accordingly, precise "targeted suppression" of each disturbance source was achieved. This adaptive suppression strategy based on spectral characteristics not only avoids excessive attenuation of the useful signal by traditional single-channel filtering, but also adjusts the filter parameters in real time according to the dynamic changes in the disturbance frequency, ensuring optimal suppression performance under complex operating conditions. Simultaneously, the phase reversal design of the suppressed signal effectively counteracts the impact of the original disturbance, while the adaptive adjustment of the amplitude balances suppression effectiveness and control stability. This provides crucial theoretical support for solving the smoothness issues such as impact and jitter caused by multi-source disturbance coupling during the braking process of the main trailer, filling the theoretical gap in the existing main trailer collaborative control in terms of dynamic disturbance suppression and smoothness coordination.

[0040] Furthermore, the modification of the output contract of each agent based on the suppression signal includes: Suppress signal Original contract instructions superimposed on the corresponding intelligent agent u i Above, the power output contract is amended to form the final instruction. The data is distributed to each braking agent. Simultaneously, the virtual disturbance observer adjusts the data based on the applied suppression... The updated estimate forms a closed-loop optimization. This method achieves coordinated suppression of synthetic perturbations through "frequency band segmentation, targeting, and adaptation."

[0041] In this embodiment, suppression signal injection and closed-loop update are implemented. By dynamically superimposing the suppression signal onto the agent's original contractual instructions, a real-time closed-loop adjustment mechanism of "perturbation-observation-suppression-feedback" is constructed. When the suppression signal... After injection, the braking agent executes the revised final command. longitudinal acceleration of the vehicle As the disturbance changes, the virtual disturbance observer immediately captures this feedback information, re-estimates the amplitude, phase, and frequency characteristics of the disturbance, and then iteratively optimizes the suppression signal parameters. This closed-loop design ensures that the suppression strategy can continuously adapt to the dynamic evolution of the disturbance. For example, the increased disturbance caused by a sudden change in the road adhesion coefficient during emergency braking, or the drift in disturbance characteristics caused by brake system thermal fade in long downhill conditions, can all be dynamically calibrated through real-time closed-loop updates to achieve the suppression effect. Compared with open-loop suppression methods, this closed-loop mechanism significantly improves the robustness and accuracy of disturbance suppression, enabling the main trailer braking system to maintain stable coordinated output under complex and variable operating conditions. It effectively alleviates braking shock and vehicle vibration caused by disturbance coupling, providing a feasible technical path to fill the theoretical gap in smooth coordinated control of the main trailer.

[0042] The following section describes in detail the trailer feedforward-model reference adaptive cooperative strategy based on the prediction of the master vehicle's motion trajectory, as formulated in step 4. (Refer to...) Figure 4 .

[0043] In one embodiment, the central decision-maker generates a predicted trajectory for the tractor unit based on the final contract instructions, including: Based on the current vehicle state and the final contractual instructions of each agent, the central collaborative decision-maker uses the vehicle's longitudinal dynamics model to generate a predicted trajectory for the tractor unit within a preset time period (e.g., 0.3-0.8 seconds). The predicted trajectory includes the predicted longitudinal deceleration of the tractor unit. (t) and desired hinge angle (t).

[0044] In this embodiment, the predicted trajectory reflects the braking strategy planned by the tractor unit to achieve smooth braking. The central decision-maker publishes this predicted trajectory to the trailer braking agent in real time, enabling the trailer to perceive the tractor unit's braking intention in advance, thus laying the information foundation for achieving smooth braking in synergy between the tractor and trailer. Unlike the traditional delayed response mode, this prediction-based information interaction method transforms the control timing between the tractor and trailer from a passive delayed relationship of "tractor unit action - trailer following" to an active coordinated relationship of "tractor unit intention - trailer prediction," significantly reducing the response delay of the trailer braking system.

[0045] Furthermore, the trailer braking intelligent agent performs trailer feedforward control and model reference adaptive feedback control, and outputs trailer braking commands, including: The trailer braking agent receives the predicted trajectory of the tractor and uses it as a reference input for the feedforward controller. Utilizing a simplified dynamic model of the trailer itself, the feedforward controller outputs the basic braking force of the trailer required to track the predicted trajectory of the tractor. ; A model reference adaptive controller is constructed, with the tractor's predicted trajectory as the reference model, the actual motion state of the trailer as the feedback, and the adaptive law minimizing the tracking error and identifying and compensating for nonlinear parameters (such as the equivalent friction coefficient) at the connection between the tractor and trailer online. The output is the feedback correction force. ; The trailer braking intelligent system combines the feedback correction force with the trailer's basic braking force: Receive trailer braking command .

[0046] Furthermore, the actual motion state of the trailer includes the measured longitudinal deceleration of the trailer and the measured articulation angle, and the tracking error includes the error between the measured longitudinal deceleration of the trailer and the predicted longitudinal deceleration of the tractor, as well as the error between the measured articulation angle and the expected articulation angle.

[0047] In this embodiment, this dual-mode mechanism achieves "predictive guidance and adaptive precision tracking," greatly reducing the lag of traditional feedback control. Furthermore, by identifying and compensating for nonlinear parameters online, it effectively improves the consistency of the main trailer's motion state under complex operating conditions. Specifically, when the tractor performs braking, the trailer braking agent, based on the received predicted trajectory of the tractor, pre-plans the trailer's basic braking force through a feedforward controller. This ensures the trailer can quickly respond to the tractor's braking intention, avoiding understeer or fishtailing caused by braking lag. Simultaneously, the model reference adaptive feedback control continuously monitors the trailer's actual motion state, comparing the measured longitudinal deceleration and articulation angle with the target values ​​in the reference model in real time. If a deviation occurs, the adaptive law immediately adjusts the feedback correction force, dynamically compensating for nonlinear disturbances caused by changes in the friction characteristics of the connecting device and uneven load distribution. This ensures the trailer always closely follows the tractor's trajectory, achieving dynamic and smooth coordination between the main trailer and the trailer.

[0048] Secondly, embodiments of this application also provide a commercial vehicle driver-trailer coordinated braking control system for executing the commercial vehicle driver-trailer coordinated braking control method as described in any embodiment of the first aspect, referring to... Figure 1 The system includes a central collaborative decision-maker, a cluster of braking execution agents, a perception layer, and a network communication layer. The central collaborative decision-maker is used to run a multi-agent collaborative optimization algorithm and output trailer braking commands. The perception layer is used to collect vehicle status information, road environment information, and driver intention input information. The network communication layer is used to provide a communication network. The braking execution intelligent agent cluster includes an EBS intelligent agent, a retarder intelligent agent, and a trailer brake intelligent agent. The EBS intelligent agent is located in the EBS controller and is used for local pressure closed-loop control, as well as for evaluating and reporting its own real-time status and capability boundaries. The retarder intelligent agent is located in the retarder controller and is used for local torque control, as well as for evaluating and reporting thermal status and torque fluctuation characteristics. The trailer brake intelligent agent is located in the trailer EBS gateway or controller and is used to receive trailer braking commands and control trailer braking.

[0049] The embodiments provided by the present invention have the following advantages: Architectural breakthroughs enhance system robustness: Through a multi-agent negotiation mechanism, dynamic and flexible allocation of braking tasks is achieved, enabling the system to adapt to changes in actuator state, fundamentally avoiding conflicts between instructions and capabilities, and significantly improving the smoothness and robustness of control under complex working conditions. Overcoming the challenge of composite vibration suppression: The proposed virtual sensing and multi-channel collaborative suppression method can effectively address the problem of disturbance synthesis from different physical sources and different frequency bands, achieving a vehicle-level "ultra-smooth" braking feel that traditional methods cannot achieve, especially improving the smoothness on low-adhesion roads or during aggressive driving. Achieving forward-looking integrated control of master and trailer: Based on the prediction-based trailer coordination method, the trailer changes from a passive and lagging response to an active and synchronous cooperation, which basically eliminates the longitudinal impulse and lateral sway of the master and trailer caused by coordination delay, and improves the driving stability and comfort of the entire vehicle convoy. A complete technical closed loop is formed: by organically combining the three key links of negotiation allocation, disturbance suppression and motion coordination, a comprehensive solution for smooth braking with complete theory, logical consistency and engineering implementation is formed, which has significant practical value.

[0050] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A commercial vehicle primary hang-on coordinated brake control method, characterized by, The method is based on a commercial vehicle driver-trailer coordinated braking control system, which includes a central coordinated decision-maker, an EBS intelligent agent, a retarder intelligent agent, and a trailer braking intelligent agent; the method includes: Obtain the overall vehicle status and total requirements; The central collaborative decision-maker decomposes tasks according to the overall demand and assigns tasks to each intelligent agent. Each intelligent agent conducts capability assessment and bidding based on the overall vehicle status. The central collaborative decision-maker runs the contract network protocol according to the bidding content to determine the output contract of each intelligent agent. A disturbance model library is constructed, and each disturbance source is estimated based on a virtual disturbance observer. The main disturbance components are identified, and suppression signals are generated based on the main disturbance components. The output contracts of each agent are corrected based on the suppression signals to obtain the final contract instructions of each agent. The central decision-maker generates a predicted trajectory for the tractor unit based on the final contract instructions, and sends the predicted trajectory to the trailer braking agent. The trailer braking agent performs trailer feedforward control and model reference adaptive feedback control, outputs trailer braking commands, and controls the trailer EBS to execute.

2. The commercial vehicle primary brake coordination control method of claim 1, characterized by, The central collaborative decision-maker decomposes tasks according to the overall requirements and assigns tasks to various intelligent agents. Each intelligent agent performs capability assessment and bidding based on the overall vehicle status, including: Based on the overall braking force requirements and vehicle ride comfort goals, the central decision-maker decomposes the braking task into multiple sub-tasks. Each sub-task has service quality requirements, and the central decision-maker broadcasts the announcements of each sub-task to each intelligent agent. The EBS agent receives announcements from each subtask, assesses the current cylinder pressure, valve body temperature, and available accessory power based on its current dynamic state, calculates the maximum or minimum pressure change curve that it can stably provide in the next decision cycle, and includes a confidence index. The retarder agent receives announcements from each subtask, assesses the current rotor temperature, speed and cooling conditions based on its current dynamic state, calculates the torque curve it provides smoothly over a future decision cycle, and includes the spectral characteristics of its torque fluctuations. The EBS agent and the retarder agent each bid on the computational content.

3. The commercial vehicle primary brake coordination control method of claim 2, characterized by The central collaborative decision-making unit operates the contract network protocol according to the bidding content to determine the output contracts of each intelligent agent, including: After receiving all bids, the central decision-maker runs a collaborative planning algorithm. The collaborative planning algorithm aims to minimize the overall impact and balance the load of each agent. Under the premise of meeting the service quality requirements of each sub-task, it generates an output contract for each winning agent. The output contract includes the agent's target output curve and corresponding reward and penalty clauses for a future decision cycle. The central decision-making unit and the winning intelligent agent signed a power contribution contract.

4. The commercial vehicle driver-trailer coordinated braking control method according to claim 1, characterized in that, The construction of the disturbance model library, which estimates each disturbance source based on a virtual disturbance observer, includes: A parameterized disturbance model library is established for EBS, retarder, and transmission system as disturbance sources; The virtual disturbance observer uses the longitudinal acceleration sensor signal of the whole vehicle as the total disturbance output, and combines it with the current contractual instructions of each agent. Through a parallel adaptive Kalman filter, it estimates the amplitude, frequency and phase of each disturbance source in real time.

5. The commercial vehicle driver-trailer coordinated braking control method according to claim 4, characterized in that, The process of identifying the main disturbance components and generating a suppression signal based on the main disturbance components includes: Real-time spectrum analysis is performed on the estimated amplitude, frequency and phase of each disturbance source to identify the main disturbance components, which include disturbance amplitude and disturbance frequency. The suppression decision module dynamically configures a set of multi-channel adaptive notch filters based on the perturbation frequency distribution. Each channel of the adaptive notch filter corresponds to a perturbation frequency. Each channel calculates the suppression signal that needs to be injected into the control command of the agent corresponding to that channel. The phase of the suppression signal is opposite to the estimated phase, and the amplitude of the suppression signal is adaptively adjusted according to the disturbance amplitude and the controller gain.

6. The commercial vehicle driver-trailer coordinated braking control method according to claim 1, characterized in that, The modification of the output contract of each agent based on the suppression signal includes: The suppression signal is superimposed on the original contract instruction of the corresponding intelligent agent to modify the output contract.

7. The commercial vehicle driver-trailer coordinated braking control method according to claim 1, characterized in that, The central decision-maker generates a predicted trajectory for the tractor unit based on the final contract instructions, including: Based on the current vehicle status and the final contract instructions of each agent, the central collaborative decision-maker uses the vehicle's longitudinal dynamics model to generate the tractor's predicted trajectory within a preset time period. The tractor's predicted trajectory includes the predicted longitudinal deceleration and the desired articulation angle.

8. The commercial vehicle driver-trailer coordinated braking control method according to claim 7, characterized in that, The trailer braking agent performs trailer feedforward control and model reference adaptive feedback control, and outputs trailer braking commands, including: The trailer braking agent receives the predicted trajectory of the tractor and uses it as a reference input for the feedforward controller. Using the simplified dynamic model of the trailer itself, the feedforward controller outputs the basic braking force of the trailer required to track the predicted trajectory of the tractor. A model reference adaptive controller is constructed. The reference model of the model reference adaptive controller is the predicted trajectory of the tractor, the feedback is the actual motion state of the trailer, the adaptive law is to minimize the tracking error and identify and compensate the nonlinear parameters at the connection between the tractor and the trailer online, and the output is the feedback correction force. The trailer braking intelligent agent combines the feedback correction force with the trailer's basic braking force to obtain the trailer braking command.

9. The commercial vehicle driver-trailer coordinated braking control method according to claim 8, characterized in that, The actual motion state of the trailer includes the measured longitudinal deceleration of the trailer and the measured articulation angle. The tracking error includes the error between the measured longitudinal deceleration of the trailer and the predicted longitudinal deceleration of the tractor, as well as the error between the measured articulation angle and the expected articulation angle.

10. A commercial vehicle driver-trailer coordinated braking control system, used to execute the commercial vehicle driver-trailer coordinated braking control method as described in any one of claims 1-9, characterized in that, The system includes a central collaborative decision-maker, a cluster of braking execution agents, a perception layer, and a network communication layer. The central collaborative decision-maker is used to run a multi-agent collaborative optimization algorithm and output trailer braking commands. The perception layer is used to collect vehicle status information, road environment information, and driver intention input information. The network communication layer is used to provide a communication network. The braking execution intelligent agent cluster includes an EBS intelligent agent, a retarder intelligent agent, and a trailer brake intelligent agent. The EBS intelligent agent is located in the EBS controller and is used for local pressure closed-loop control, as well as for evaluating and reporting its own real-time status and capability boundaries. The retarder intelligent agent is located in the retarder controller and is used for local torque control, as well as for evaluating and reporting thermal status and torque fluctuation characteristics. The trailer brake intelligent agent is located in the trailer EBS gateway or controller and is used to receive trailer braking commands and control trailer braking.

Images