Commercial vehicle free cruise control method and system, computer equipment and storage medium

Abstract

The invention discloses a free cruise control method and system for a commercial vehicle, computer equipment and a storage medium, and relates to the technical field of automatic driving and intelligent traffic systems, and the method comprises the steps: building a container truck driving environment potential field model; designing a prediction controller according to the container truck free cruise strategy model; model predictive control stipulates a target function, and control input is solved for target control; and generating a predictive control track to interact with a distributed controller of a surrounding intelligent vehicle. The method has the capability of adapting to a complex mixed traffic environment, can dynamically and flexibly carry out multi-objective optimization control, and comprises comprehensive integration of multiple constraint conditions, efficient cooperative control and blind area elimination.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic driving and intelligent transportation systems, and specifically relates to a commercial vehicle free cruise control method, system, computer equipment and storage medium. Background Art

[0002] Cruise control strategies for commercial vehicles have gradually evolved from traditional fixed modes to more flexible free cruise control. Free cruise control strategies not only focus on driving safety and comfort in single-vehicle environments but also address control issues in multi-vehicle coordination, complex interactions, and dynamic environments. Numerous studies and patents have proposed various free control strategies tailored to environmental changes and diverse driving scenarios, providing new solutions for cooperative control in commercial vehicle platooning and mixed traffic environments. However, these technologies still have significant shortcomings in complex environments.

[0003] Cao Xuanhao designed a robust control strategy for the longitudinal dynamics and actuator uncertainty problems of autonomous vehicles. The three-state vehicle following model and descriptor method were used to effectively decouple uncertain parameters and reduce the conservatism of controller design. Li Xingkun et al. proposed an adaptive distance domain predictive cruise control strategy for heavy-duty commercial vehicles based on intelligent distance prediction (IDP). By combining the longitudinal dynamics model of the vehicle with the traffic flow prediction model, the IDP model was used to predict the motion trajectory of the vehicle in front, and the following distance and speed were dynamically adjusted based on the prediction results. However, it did not consider that in mixed traffic environments, it may not be possible to fully predict and adapt when interacting with non-intelligent or non-communicative vehicles. Bai Chenguang proposed a commercial vehicle adaptive cruise control strategy based on fuzzy MPC (model predictive control). The lower-level control uses PID feedback control to achieve precise control of the driving torque and braking pressure. The upper-level control uses fuzzy MPC to make expected acceleration decisions based on the vehicle state, and dynamically adjusts the weight balance between following and comfort according to different driving scenarios. Huang Peng adopted a hierarchical control method that combines following and comfort. In order to achieve the goals of vehicle headway and fuel economy, an upper-level controller was designed based on the time-to-headroom model and model predictive control (MPC) theory, and the lower-level controller adopted a feedforward plus feedback strategy for acceleration control. In addition, to improve the adaptability of the system under complex working conditions, a fuzzy controller based on vehicle headway error and relative speed was introduced to adjust the weight coefficients of followability and comfort online. Mao et al. proposed an improved adaptive cruise control strategy that dynamically adjusted the headway based on the deceleration time and deceleration change of the leading vehicle to improve driving safety and road utilization. The strategy was controlled and verified by a model predictive controller. The results showed that the strategy could maintain vehicle headway more smoothly, improving driving safety and comfort. However, the adaptability and responsiveness of the strategy in complex road environments or emergency braking conditions still need to be further verified. Sun et al. proposed a semi-regular adaptive cruise decision-making strategy based on the deep deterministic policy gradient (DDPG) algorithm for heavy-duty intelligent vehicles. They established an accurate three-axis vehicle load model, calculated the load transfer rate, and performed active control to improve the roll stability of heavy-duty vehicles on high-speed curves. Although deep reinforcement learning can dynamically adjust vehicle speed and distance between vehicles to achieve better vehicle following and cornering stability, the complexity of modeling increases the difficulty of policy migration and adaptation to different vehicle types, and its safety remains to be verified.

[0004] In summary, existing free-cruise strategies for commercial vehicles have achieved certain results in various aspects, such as robust control, predictive control models, and multi-objective optimization, which have improved vehicle followability, stability, and fuel economy. However, these strategies still lack adaptability to interactions with non-intelligent vehicles in complex mixed traffic environments, modeling complexity, and dynamic response capabilities. Based on this, this project proposes a "free-cruise control method for multi-vehicle cooperative formations of commercial vehicles." By establishing a potential field model of the vehicle driving environment and combining it with a distributed model predictive control strategy to describe the complex interactive relationship between the controlled vehicle and the environment, it aims to more comprehensively eliminate control blind spots and achieve flexible switching and collaborative control of multiple control effects. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that the current free cruise control strategy mainly relies on communication between intelligent vehicles to achieve cooperative control, but in mixed traffic environments, there are a large number of non-intelligent or non-communicative vehicles. The adaptability of existing technologies in these situations is poor. When commercial vehicles interact with non-communicative traditional vehicles, the system is often unable to accurately predict and adapt to the behavioral changes of these vehicles, which may lead to dangerous situations such as emergency braking and collisions. Therefore, how to effectively coordinate control in complex traffic environments, especially with non-intelligent or non-communicative vehicles, is a major shortcoming of existing technologies.

[0007] Existing cruise control strategies typically employ pre-set, fixed control logic, which works well in specific scenarios. However, faced with evolving traffic conditions and dynamic interactions between vehicles, these traditional strategies can be slow to respond, fail, or become overly limited. For example, when road conditions suddenly change or when the need for coordinated control between multiple vehicles arises, traditional control methods often fail to adapt quickly, resulting in the system's inability to make accurate and timely decisions in complex situations. Therefore, a more flexible and comprehensive control strategy is needed that can cope with diverse traffic conditions and adapt dynamically.

[0008] Existing free cruise control strategies often focus on optimizing a single or a few objectives, such as vehicle following, comfort, and stability. However, in practice, cruise control must simultaneously meet multiple objectives, such as driving safety, energy consumption, and speed control, often conflicting with each other. Furthermore, existing control strategies lack the ability to balance and dynamically adjust the weights of multiple objectives in complex environments, resulting in an inability to provide optimal solutions in diverse driving scenarios. Therefore, optimizing and adjusting multiple objectives and their weights based on different cruise control objectives and complex scenarios is crucial to achieve the desired switching between multiple control effects and meet the requirements of hybrid dynamic control.

[0009] Existing adaptive cruise control strategies for commercial vehicles typically only address a subset of constraints, focusing primarily on factors such as vehicle followability and comfort, while ignoring the geometric and physical constraints of vehicles on real roads. Physical characteristics such as a vehicle's turning radius, acceleration and deceleration performance, braking pressure, and steering ability are crucial for safe and stable operation, yet these physical constraints are often underestimated or not effectively integrated into control strategies. Therefore, comprehensively considering multiple constraints and improving constraint handling capabilities under complex operating conditions will enhance system responsiveness and better enable free cruise control in multi-vehicle cooperative platooning.

[0010] To solve the above technical problems, the present invention provides the following technical solutions: a commercial vehicle free cruise control method, comprising: establishing a potential field model of the container truck driving environment; designing a predictive controller based on the container truck free cruise strategy model; specifying the objective function through model predictive control, solving the control input for target control; generating a predictive control trajectory to interact with the distributed controllers of surrounding intelligent vehicles.

[0011] As a preferred solution of the commercial vehicle free cruise control method of the present invention, wherein: the establishment of the container truck driving environment potential field model includes environmental potential field modeling and driving direction potential field modeling;

[0012] The environmental potential field modeling includes obstacle vehicle potential field modeling and road potential field modeling;

[0013] The driving direction potential field modeling includes providing a vehicle driving tendency toward the front of the lane, and changing lanes when an obstacle appears in front of the lane.

[0014] As a preferred solution of the commercial vehicle free cruise control method of the present invention, wherein: the obstacle vehicle potential field modeling includes using the vehicle longitudinal influence as a skeleton and extending it laterally to describe the vehicle influence distribution;

[0015] A local coordinate system is established with the vehicle body direction as the positive direction of the x-axis and the rear end point of the vehicle outer contour as the origin;

[0016] Calculate the longitudinal potential field value of the environmental vehicle in sections;

[0017] Determine the distance between the vehicle and the surrounding environment based on the relative speed;

[0018] The longitudinal potential field risk distribution is reflected according to the vehicle distance relative to the environment.

[0019] As a preferred solution of the commercial vehicle free cruise control method of the present invention, wherein: the obstacle vehicle potential field modeling further includes calculating the lateral potential field of the environment vehicle using a Gaussian-like function;

[0020] The total value of the environmental vehicle potential field is obtained by combining the longitudinal potential field value of the environmental vehicle with the transverse potential field;

[0021] Determine the location of the potential field calculation point around the vehicle and determine the lateral impact range of the potential field.

[0022] As a preferred embodiment of the commercial vehicle free cruise control method of the present invention, the road potential field modeling includes establishing lane line constraint relationships through the road potential field, including lane lines that the vehicle cannot cross and lane lines that the vehicle can cross;

[0023] When the lane line constraint relationship is that the lane line can be crossed, the potential field calculation point is set to calculate the generated road potential field;

[0024] When the lane line constraint relationship is that the vehicle cannot cross the lane line, a quadratic curve is used to describe the influence of the road boundary on the road potential field, and the lane line constraint relationship is superimposed to obtain the entire road potential field.

[0025] As a preferred embodiment of the commercial vehicle free cruise control method of the present invention, wherein: the designing of the predictive controller includes calculating the objective function by a distributed model predictive controller to optimize the vehicle's future control input;

[0026] In a fixed ground coordinate system, a predictive controller is designed by establishing a kinematic model of the container truck.

[0027] The vehicle dynamics model predicts the vehicle state over a period of time in the future and controls the inputs based on the constraints;

[0028] The vehicle kinematic model is optimized by using the constraint of increasing the control increment;

[0029] Through rolling optimization, MPC dynamically adjusts the control strategy based on the vehicle's current state and the real-time updated environmental potential field to eliminate control blind spots.

[0030] As a preferred embodiment of the commercial vehicle free cruise control method of the present invention, the objective function includes the potential field value of the driving environment potential field of the vehicle kinematic model combined with the calculation function prediction step length of the potential field model in combination with the following vehicle control, the target speed and the economic comfort requirement;

[0031] Predict the lateral and longitudinal offset errors between the predicted trajectory of the following vehicle and the preceding vehicle in the time domain, predict the deviation between the speed and the target speed in the time domain, provide a reference target speed, and achieve the minimum incremental value considering the requirements of economy and comfort;

[0032] Combined with the predicted control trajectory transmitted by the pilot vehicle, the objective function is constructed for tracking and the corresponding constraints are set.

[0033] As a preferred solution of the commercial vehicle free cruise control method of the present invention, the target control includes adding linear constraints including vehicle speed control constraints and wheel deflection angle increment constraints.

[0034] Another object of the present invention is to provide a free cruise control system for commercial vehicles, which improves the safety, efficiency and flexibility of vehicles during platoon driving. By achieving information sharing and intelligent coordination between vehicles, the system can effectively solve potential safety hazards that may arise during driving, such as the risk of collision caused by vehicles being too close or too far apart. At the same time, it also optimizes traffic flow, reduces energy consumption, and improves overall operational efficiency. In addition, the free cruise control system allows vehicles to autonomously adjust their speed and driving trajectory within a certain range, improving the adaptability and flexibility of driving, thereby effectively responding to the challenges of different road conditions and traffic environments.

[0035] To solve the above technical problems, the present invention provides the following technical solutions: a commercial vehicle free cruise control system, comprising: an environmental potential field modeling module, used to establish a potential field model of roads and environmental vehicles, describe the interaction with the surrounding environment of the controlled vehicle, and provide a basis for safety distance and risk assessment; a vehicle kinematic model module, used to generate the kinematic equations of the controlled vehicle, define the relationship between the vehicle state and input, and support dynamic simulation of the control system; a model predictive control module, used to design a controller based on the vehicle kinematic model to optimize the vehicle's travel along the target trajectory; an objective function design module, used to construct an optimization objective function, which includes constraints on the accuracy of tracking the trajectory of the vehicle ahead, speed deviation, and control increment, to ensure that the optimization result meets driving requirements; a collaborative control decision module, used to integrate information from each module, perform multi-vehicle collaborative control, and ensure the safety and stability of platoon driving.

[0036] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the method described in the configuration of a large power plant relay protection information system are implemented.

[0037] A computer-readable storage medium stores a computer program thereon, wherein when the computer program is executed by a processor, the steps of the method described in the configuration of a relay protection information system of a large power plant are implemented.

[0038] The present invention has the following beneficial effects: It is capable of adapting to complex mixed traffic environments. Existing technologies often lack effective prediction and response capabilities when interacting with non-intelligent or non-communicative vehicles. This invention, by establishing a driving environment potential field model, models and dynamically describes commercial vehicles and conventional vehicles in mixed traffic environments. This model effectively addresses the current limitations of predicting environmental complexity and non-intelligent vehicle behavior, significantly improving the collaborative control capabilities of commercial vehicle platoons in complex environments, particularly their safety and stability during dynamic interactions.

[0039] This invention innovatively introduces a distributed model predictive control strategy to achieve dynamic adjustment and multi-objective optimization of vehicle cruise control. Compared to traditional fixed-logic control methods, this technology can dynamically balance multiple objectives, such as safety, comfort, fuel economy, and vehicle responsiveness, based on real-time traffic changes. This technology can rapidly adapt to emergency braking, lane changes, or complex platooning scenarios, significantly improving the flexibility and overall performance of the cruise control system.

[0040] To address the problem that traditional free cruise control methods lack consideration of physical and geometric constraints, this patent organically integrates the vehicle's dynamic characteristics (acceleration and deceleration performance, braking pressure, turning radius, etc.) with environmental constraints (such as road boundaries and obstacles) into the control strategy. This approach ensures vehicle driving safety and stability in complex scenarios while optimizing space utilization during multi-vehicle collaboration.

[0041] By combining environmental potential field models with distributed control methods, this technology accurately describes the dynamic interaction between vehicles and their surroundings in multi-vehicle cooperative platooning, eliminating the blind spots in traditional control systems. This makes cooperative control between vehicles more efficient, enabling both efficient platooning and flexible free cruising while maintaining stability and safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive efforts. Among them:

[0043] Figure 1 This is a flowchart of an operation identification method for a commercial vehicle free cruise control method provided by one embodiment of the present invention.

[0044] Figure 2 A structural diagram of an artificial potential field model of a road environment for a commercial vehicle free cruise control method provided by one embodiment of the present invention.

[0045] Figure 3 A schematic diagram of the longitudinal potential field of an ambient vehicle in a free cruise control method for a commercial vehicle provided by one embodiment of the present invention.

[0046] Figure 4 Schematic diagram of the longitudinal and transverse directions of the vehicle environment of a commercial vehicle free cruise control method provided by one embodiment of the present invention.

[0047] Figure 5 A schematic diagram of a lane line potential field for a free cruise control method for a commercial vehicle provided by one embodiment of the present invention.

[0048] Figure 6 A vehicle kinematic model diagram of a free cruise control method for a commercial vehicle provided by one embodiment of the present invention.

[0049] Figure 7 A system functional architecture diagram of a commercial vehicle free cruise control system provided by one embodiment of the present invention.

[0050] Figure 8 The present invention provides an example schematic diagram of a commercial vehicle free cruise control method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without making creative efforts should fall within the scope of protection of the present invention.

[0052] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0053] Example 1

[0054] Reference Figures 1-6 , which is the first embodiment of the present invention, provides a commercial vehicle free cruise control method, comprising:

[0055] To address the shortcomings of existing research, which lacks consideration of the adaptability of interactions with non-intelligent vehicles in complex mixed traffic environments, modeling complexity, and system dynamic response capabilities, this project proposes a control strategy. First, by building a potential field model of the vehicle's driving environment, this strategy directly describes the complex interactions between the controlled vehicle and the environment, rather than focusing solely on the interaction logic in a single scenario. This provides a comprehensive reference for the controller's final control decisions. Next, a distributed model predictive approach is employed to complete the controller design. Because model predictive control can handle multiple objectives and constraints, it aligns with the application scenario of this system. Furthermore, model predictive control involves designing a specific objective function based on the requirements, solving the objective function to obtain control inputs, and ultimately achieving the control objectives. This approach also aligns with the logic of hierarchical control. Upper-level controllers can directly influence lower-level controllers by changing the corresponding parameters or weights of the objective function, ultimately achieving a switching effect among multiple control effects and meeting the requirements of hybrid dynamic control. Predictive control, combined with the potential field model, maximizes the use of environmental information acquired by the vehicle to eliminate control blind spots. Furthermore, the generated predictive control trajectory can interact with the distributed controllers of other surrounding intelligent vehicles to achieve coordinated control.

[0056] Step S1: Establish a potential field model of the truck driving environment.

[0057] Furthermore, by establishing an artificial potential field model of the environment, the goal is to fully describe the interactive effects of the controlled vehicle's surrounding environment on it. Through model calculations, the distribution of hazards in the space where the controlled vehicle is located can be described as the corresponding potential field value, thereby providing a basis for the vehicle's collaborative control decision-making. The factors that affect the vehicle while it is driving on an ideal road mainly include other vehicles in the environment, roads, and its own control targets. The structure of the driving environment potential field model designed in this project is as follows Figure 2 shown.

[0058] It should be noted that the establishment of the container truck driving environment potential field model includes environmental potential field modeling and driving direction potential field modeling;

[0059] For a controlled vehicle, the environmental artificial potential field u mn The calculation consists of the following parts:

[0060] u mn =r mn +e mn +s mn

[0061] Among them, the road potential field value at this point is r mn , the potential field value of the environmental vehicle is e mn , the target velocity potential field value of the vehicle is s mn .

[0062] Specifically, environmental potential field modeling includes obstacle vehicle potential field modeling and road potential field modeling;

[0063] It should be noted that in obstacle vehicle potential field modeling, the purpose of constructing an environmental vehicle potential field is to describe the impact of surrounding vehicles (i.e., environmental vehicles) on the controlled vehicle. This aims to maintain a safe distance between the controlled vehicle and the environmental vehicles based on their respective speeds. For environmental vehicles, given their motion characteristics and the actual conditions of road travel, the level of danger in their vicinity is unevenly distributed in the longitudinal and lateral directions. For example, the primary factor affecting vehicle safety in the lateral direction is the distance between vehicles, while in the longitudinal direction, speed is also a significant factor, in addition to distance. Therefore, when modeling the environmental vehicle potential field, the model design also incorporates these different characteristics in the longitudinal and lateral directions.

[0064] Due to the characteristics of vehicles driving on the road, the longitudinal influence range of each other is relatively larger than the lateral influence range. The characteristics of the creation of the environmental vehicle potential field in this paper are: using the longitudinal influence of the vehicle as the skeleton, extending it laterally to describe the distribution of vehicle influence.

[0065] A of the longitudinal potential field of the ambient vehicle car Calculate as Figure 3 As shown in the figure, a local coordinate system is established with the direction of the vehicle body as the positive direction of the x-axis and the end point of the rear end of the vehicle's outer contour as the origin.

[0066] Figure 3 The bold black part in the figure is the "skeleton" distribution of the vehicle potential field. The entire longitudinal potential field distribution is a piecewise function, which calculates the longitudinal potential field value of the surrounding vehicle piecewise. The calculation rules are as follows:

[0067]

[0068] Among them, U car is the vehicle potential field constant, representing the maximum value of the ambient vehicle potential field; v ris the relative speed between the controlled vehicle and the environment vehicle. When the speeds of the two vehicles are in the same direction, the speed of the controlled vehicle is greater than that of the environment vehicle v r >0, otherwise v r <0; K is the longitudinal distance between the controlled vehicle and the surrounding vehicle; S is a set safety distance, where S = v r ·ΔT+s min ,ΔT is the system delay, which is related to sensor delay and calculation delay, s min It is a set extended safety distance.

[0069] When point p is in the β region, such as Figure 4 In p2, the longitudinal potential field distribution is a constant U car , when point p is in the α region, as shown in p1 in the figure.

[0070] Consider the relative speed v between the environment vehicle and the controlled vehicle r , when v r When ≤0, it means that the relative distance between the environmental vehicle and the controlled vehicle gradually increases, and the longitudinal potential field value takes 0.

[0071] When v r When it is greater than 0, it indicates that the distance between the controlled vehicle and the surrounding vehicles is gradually decreasing. The closer to the controlled vehicle, the more dangerous it is, and the potential field value is negatively correlated with the relative distance; the greater the relative speed, the more dangerous it is, and the potential field value is positively correlated with the relative speed.

[0072] The above longitudinal potential field distribution reflects the risk distribution of the vehicle in the longitudinal direction and the impact of the vehicle on the surrounding environment in the lateral direction. The overall potential field is formed by extending the longitudinal direction. The specific calculation is as follows: Figure 4 shown.

[0073] The calculation of the environmental vehicle potential field is based on the longitudinal potential field coefficient A car The lateral potential field is obtained by multiplying it by the lateral potential field. The lateral potential field of the ambient vehicle is calculated using a Gaussian-like function. The total value of the ambient vehicle potential field is obtained based on the longitudinal potential field value of the ambient vehicle combined with the lateral potential field. The overall calculation method is as follows:

[0074]

[0075] Where D is the equivalent distance of the lateral potential field, and when p is in front of the vehicle, it is the Euclidean distance between the potential field calculation point around the environment vehicle and the edge of the environment vehicle, as shown in Figure 4 As shown in , when the calculation point is p3, D = p3;

[0076] When the p calculation point is on the vehicle side p2, D = p2;

[0077] When the calculation point p is at p1 behind the vehicle, D=p1,σ vis the convergence coefficient of the vehicle potential field, which determines the lateral influence range of the potential field.

[0078] It should also be noted that when a controlled vehicle is driving along a lane or changing lanes, it must consider the lane line constraints on the road. These constraints can be described by the road potential field. The lane line constraints established through the road potential field include the road boundary lane lines, which are the lane lines on both sides that the vehicle cannot cross, and the crossable lane lines, which means that the vehicle can cross and change lanes according to actual conditions.

[0079] like Figure 5 As shown in the figure, when the lane line constraint is that the lane line can be crossed, the controlled vehicle is driving in the middle lane. Then the two lane lines l2 and l3 in the middle of the road generate the road potential field for the potential field calculation point (m,n):

[0080]

[0081] Among them, U lane is the lane line potential field coefficient d lane,i is the distance l from a lane line in section A i distance; σ r is the lane line potential field convergence coefficient;

[0082] k(i) is the lane line gain coefficient, and its value is related to the location of the controlled vehicle. When the controlled vehicle is traveling in the left lane, the adjacent lane line to the controlled vehicle is l2, so k(2) = 1. Since l3 is separated from the controlled vehicle by a lane, we believe that during driving, the road constraint on vehicles is to keep each vehicle in its own lane and avoid large-scale lane changes as much as possible. Therefore, k(3) is greater than k(2). In this paper, k(3) = 2.

[0083] When the lane line constraint relationship is that the vehicle cannot cross the lane line, a quadratic curve is used to describe the influence of the road boundary on the road potential field, and the lane line constraint relationship is superimposed to obtain the entire road potential field.

[0084] like Figure 5 As shown in the figure, the road boundary lines l1 and l4 are insurmountable, and the influence of the road boundary on the road potential field gradually decreases to zero on the side facing the lane and continues to increase to infinity on the side facing away from the lane. This paper uses a quadratic curve to describe this characteristic, and the calculation formula is as follows:

[0085]

[0086] The entire road potential field r mn The calculation formula is the superposition of the middle lane line of the road and the boundary lines on both sides of the road:

[0087]

[0088] Furthermore, the driving direction potential field modeling of the embodiment of the present invention provides a vehicle's driving tendency toward the front of the lane, and when an obstacle appears in front of the lane, the vehicle changes lanes.

[0089]

[0090] Among them, x r is the longitudinal distance between the potential field point and the controlled vehicle, The velocity potential field coefficient is a negative constant, and ε is a positive constant, which ensures that the potential field value is always positive in the prediction time domain.

[0091] Ignoring the influence of other environmental factors, the potential field function makes the potential field in front of the controlled vehicle lower than the potential field behind it, so that the controlled vehicle maintains a tendency to move forward.

[0092] Step S2: Design a predictive controller based on the container truck free cruising strategy model.

[0093] like Figure 6 As shown, preferably, the objective function is calculated by a distributed model predictive controller to optimize the future control input of the vehicle;

[0094] In the ground-fixed coordinate system OXY, the vehicle kinematic equation is:

[0095]

[0096] Among them, (x, y) is the coordinate of the center of the vehicle's rear axle, is the vehicle heading angle, δ is the vehicle front wheel deviation angle, v is the vehicle rear axle speed, l is the wheelbase, then the system can be regarded as an input u=[v,δ] T , the state quantity is control system.

[0097] Assuming Δt is a step length of discrete time, the discretized vehicle kinematic model is as follows:

[0098]

[0099] Where χ(k) represents the state of the system at the kth step.

[0100] In order to increase the constraints on the control increment, the discretized vehicle kinematic model needs to be transformed. Let Δu(k) = u(k) - u(k-1), then the discretized vehicle kinematic model is transformed into,

[0101]

[0102] Among them, ξ(k+1) represents the state variable of the system at the k+1th step, Δv and Δδ represent the change in the rear axle velocity and the front wheel angle of the vehicle within two discrete time steps, respectively.

[0103] Through rolling optimization, MPC dynamically adjusts the control strategy based on the vehicle's current state and the real-time updated environmental potential field to eliminate control blind spots.

[0104] Step S3: The model predictive control specifies the objective function and solves the control input to perform target control.

[0105] Furthermore, the following vehicle control combines the environmental vehicle potential field, target speed, and economic comfort requirements, and combines the vehicle kinematic model with the calculation function of the potential field model to predict the potential field value of the driving environment potential field with the step length;

[0106] Combined with the predicted control trajectory transmitted by the pilot vehicle, the objective function is constructed for tracking and the corresponding constraints are set;

[0107] In addition to the environmental vehicle potential field, target speed, and economic comfort requirements, the following vehicle control also needs to track the predicted control trajectory transmitted by the pilot vehicle. Therefore, the objective function is constructed as follows

[0108]

[0109] Among them, U APF (i) The driving environment potential field for each prediction step. By substituting the vehicle kinematic model mentioned above into the potential field calculation function, the potential field value of the controlled vehicle position at each prediction step i in the prediction time domain is obtained. This is used to evaluate the risk of the controlled vehicle position on the predicted trajectory. The control trajectory of the autonomous vehicle depends largely on the calculation of the safety potential field function.

[0110] e v (i)=v(i)-v des To predict the deviation between the speed and the target speed in the time domain, this item mainly provides a reference target speed for the vehicle in an obstacle-free situation; Δu(k+i|t) is the control increment. This addition can minimize the increment when optimizing the solution, increase the comfort of the vehicle and improve fuel economy; μ is the relaxation factor, which is set to prevent the optimization solution from having no solution; Q, R, and P are weight matrices, and N p is the prediction time domain, N c To control the time domain, v des A reference speed is set.

[0111] It should be noted that in order to ensure that the optimization results are consistent with the actual situation, linear constraints are added, including vehicle speed control constraints and wheel angle increment constraints.

[0112] The objective function uses the control increment as input, and the constraints on the control quantity itself need to add linear constraints. In order to make the unmanned vehicle's cruising process smoother, the vehicle's speed control constraint is set to:

[0113] 0≤v≤1.1*v des

[0114] -4.9m / s 2 ≤a≤1m / s 2

[0115] Among them, v des is the desired speed of the vehicle, and a is the vehicle acceleration.

[0116] In lateral control, the vehicle steering wheel takes 1.8 seconds to turn one circle, corresponding to a front wheel deflection angle of 17°. The wheel deflection angle increment constraint is set as follows:

[0117] -25°≤δ≤25°

[0118] -9.4° / s≤ω≤9.4° / s

[0119] Where δ is the front wheel deflection angle of the vehicle, and ω is the rotational acceleration of the front wheel deflection angle of the vehicle.

[0120] Step S4: Generate a predicted control trajectory to interact with the distributed controllers of surrounding intelligent vehicles.

[0121] Specifically, a decentralized blockchain network is established, with all participating vehicles connected to the network as nodes. Each node saves a copy of the information to ensure data security and transparency.

[0122] Through efficient real-time communication protocols, including MQTT or WebSocket, to reduce latency and support fast data transmission, each smart vehicle is equipped with multiple sensors, including radar, cameras, lidar, etc., to collect data on the surrounding environment in real time, including the location, speed, driving direction and road conditions of other vehicles; the internal system of the vehicle performs preliminary processing on the collected data, extracts important features, including dangerous conditions, traffic signal changes, etc., and formats the data to prepare for subsequent transmission.

[0123] Before data is uploaded to the blockchain, it is encrypted and digitally signed to prevent tampering and ensure the authenticity of the information source. The processed data is then written to the blockchain network according to the set block structure. A transaction confirmation mechanism is set up to ensure the timeliness of information, such as through network consensus algorithms such as PoW or PoS to verify the validity of the data.

[0124] Real-time data broadcasting is achieved through the set communication protocol. When a vehicle detects information that needs to be shared, it immediately disseminates the data to other surrounding vehicles through the blockchain or distributed network. After receiving the broadcast information, other vehicles perform decryption processing and digital signature verification to confirm the authenticity and integrity of the information.

[0125] After receiving information from surrounding vehicles, the intelligent decision-making system inside the vehicle uses an information fusion algorithm to combine the vehicle's sensor data with the received external information to improve its perception of the environment. Based on the fused information, the vehicle dynamically generates driving strategies, such as changing lanes, slowing down or accelerating, to adapt to complex traffic environments. After executing driving decisions, the vehicle continuously monitors changes in vehicle performance and the surrounding environment, uploads new data to the chain again, and updates the information pool in real time.

[0126] Example 2

[0127] Reference Figure 7 , is an embodiment of the present invention, which provides a commercial vehicle free cruise control system, including: an environmental potential field modeling module 100, a vehicle kinematic model module 200, a model predictive control module 300, an objective function design module 400, and a collaborative control decision module 500.

[0128] The environmental potential field modeling module 100 is used to establish a potential field model of the road and the environmental vehicle, describe the interaction with the surrounding environment of the controlled vehicle, and provide a basis for safety distance and risk assessment.

[0129] Specifically, it is responsible for establishing a potential field model of the vehicle's driving environment to reflect the impact of the surrounding environment on the controlled vehicle, including other vehicles, road constraints, etc.; collecting surrounding environment information in real time through sensors (such as radar, lidar, cameras, etc.), including vehicle position, speed, driving direction, road status, etc.; calculating the environmental vehicle potential field, road potential field and target speed potential field based on sensor data.

[0130] The calculation results include:

[0131] Environmental vehicle potential field value: defines the safe distance and speed relationship between the vehicle and other surrounding vehicles.

[0132] Road potential field value: The potential field value calculated based on road constraints (such as lane lines).

[0133] The calculated potential field data are output for use by subsequent modules to ensure that these data are updated in real time.

[0134] Real-time updated potential field data effectively reduces the risk of collision with surrounding vehicles, ensures safe driving of vehicles in different environments, and enables vehicles to maintain an appropriate safety distance under complex traffic conditions, thereby improving road safety and driving stability.

[0135] The vehicle kinematics model module 200 is used to generate the kinematic equations of the controlled vehicle, define the relationship between the vehicle state and input, and support dynamic simulation of the control system.

[0136] Specifically, a kinematic model of the controlled vehicle is provided to describe the dynamic characteristics of the vehicle under the influence of the environmental potential field; real-time potential field information, especially the environmental vehicle and road potential fields, is received from the environmental potential field modeling module; based on the vehicle's motion characteristics and potential field information, a kinematic model of the controlled vehicle is established to describe the vehicle's state, including position, speed, heading angle, etc.

[0137] The current vehicle state and predicted future state information are output to the model predictive control module to support the generation of control inputs. The control system can quickly respond to changes in vehicle state, improving control precision. Through dynamic simulation, the vehicle can better adapt to different driving conditions, ensuring stable driving in various environments, thereby improving driving safety and passenger comfort.

[0138] The model predictive control module 300 is used to design a controller to optimize the vehicle's travel along a target trajectory based on the vehicle's kinematic model.

[0139] Specifically, model predictive control (MPC) is implemented to generate control inputs based on the current state and environmental potential field information.

[0140] The current state of the controlled vehicle and its prediction are obtained from the vehicle kinematic model module. The MPC target is constructed according to the objective function information and constraints provided by the objective function design module. The MPC algorithm is used to perform rolling optimization based on the current state, potential field information and objective function, and the next control input is calculated. The calculated control input is output to the objective function design module to achieve target control.

[0141] Specifically, by implementing Model Predictive Control (MPC), the system is able to optimize vehicle driving paths in complex and dynamically changing environments. MPC's rolling optimization capability enables the vehicle to adjust its driving strategy in real time, effectively reducing the incidence of traffic accidents and improving vehicle controllability and responsiveness in complex traffic conditions.

[0142] The objective function design module 400 is used to construct an optimization objective function, which includes constraints on the accuracy of tracking the trajectory of the vehicle ahead, speed deviation, and control increment, to ensure that the optimization result meets driving requirements.

[0143] Specifically, it is responsible for designing the objective function to meet the vehicle's motion goals and safety constraints; obtaining the state and potential field information related to the vehicle and the environment from the vehicle kinematic model module, constructing the objective function, and outputting the objective function and its constraints to the model predictive control module for it to optimize the control input; ensuring that the vehicle can accurately track the target trajectory during driving, reducing speed deviation during driving, ensuring the stability and safety of the system, and effectively improving driving safety and comfort, so that the vehicle can maintain optimal performance at different speeds and driving conditions.

[0144] The collaborative control decision module 500 is used to integrate information from various modules and perform multi-vehicle collaborative control to ensure the safety and stability of platoon driving.

[0145] Specifically, it is responsible for formulating collaborative control strategies to ensure coordinated operations with surrounding smart vehicles; and receiving information shared by surrounding smart vehicles in real time through blockchain or distributed networks, including speed, location, driving direction, etc.

[0146] The system integrates the vehicle's sensor data with the received external information to improve the overall environmental perception capability. Based on the fused information, it generates driving strategies that adapt to complex traffic environments, such as lane changing, deceleration, acceleration, etc. The collaborative control decisions are fed back to the vehicle kinematic model module to update the actual state of the vehicle and adjust future control.

[0147] By integrating information sharing between multiple vehicles, this module can effectively improve the efficiency and safety of overall traffic flow. The implementation of collaborative control strategies ensures the safety of vehicles in platooning while optimizing the efficiency of the entire fleet, reducing overall traffic delays and improving road utilization.

[0148] Example 3

[0149] The third embodiment of the present invention is different from the first two embodiments in that:

[0150] Specifically, if the commercial vehicle free cruise control method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program codes.

[0151] The computer program includes several instructions that can cause a computer device (such as a server, a personal computer or other network device) to perform the following steps:

[0152] Establish environmental potential field and driving direction potential field models, covering obstacles and road conditions. By using potential field values ​​to represent the risks of the surrounding environment, ensure that the vehicle maintains a safe distance from other vehicles, determine the calculation rules of the longitudinal and lateral potential fields, and model different environmental factors (such as the distance between vehicles and relative speed).

[0153] The application of model predictive control (MPC) method allows handling of dynamic environments with multiple objectives and constraints; vehicle motion and control inputs are optimized by discretizing the vehicle kinematic model; and control strategies are dynamically adjusted using real-time environmental information to eliminate control blind spots.

[0154] The objective function is constructed by combining the environmental potential field, target speed, and comfort requirements. Constraints are set for the predicted path to improve the safety and comfort of the vehicle speed. The effectiveness of the optimization results is ensured by introducing linear constraints to reflect the actual driving situation.

[0155] Design a decentralized blockchain network to achieve secure and transparent data sharing; use efficient real-time communication protocols (such as MQTT or WebSocket) for data transmission, and equip it with multiple sensors to obtain surrounding environment information; encrypt and format the acquired data to ensure the authenticity and integrity of the information.

[0156] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0157] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0158] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0159] The computer program of the present invention further supports real-time operations. For example, the computer program of the present invention realizes real-time response and control of complex traffic environments by establishing a dynamic potential field model and distributed model predictive control (MPC). Vehicles collect surrounding data in real time through sensors, and use a decentralized blockchain network to ensure information security and transparency, and quickly share information on environmental changes. Compared with traditional detection methods, the information fusion algorithm improves the ability to perceive the environment, allowing the vehicle to generate and adjust driving strategies in real time. At the same time, continuous monitoring and feedback loop mechanisms ensure the effectiveness of each decision, enabling the control system to achieve safe and flexible self-driving functions in complex mixed traffic environments.

[0160] Example 4

[0161] Reference Figure 8 , which is an embodiment of the present invention, provides a commercial vehicle free cruise control method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0162] During the free cruising of an autonomous container truck, the vehicle must make real-time decisions based on the surrounding environment (including traffic flow, obstacles, etc.). Traditional decision-making methods are often limited to specific scenarios. To better cope with complex and changing road environments, this paper proposes a dynamic model predictive control method based on an environmental potential field model. Through potential field calculations, the container truck's driving strategy is adjusted in real time, thereby optimizing the driving path and avoiding potential traffic accidents.

[0163] Specific implementation steps:

[0164] Vehicle driving environment potential field modeling

[0165] In this embodiment, during driving, the container truck is considered as a controlled object located in the middle lane. How to interact with other vehicles traveling in the same direction to improve driving safety is a problem considered by the present invention. Based on the environmental potential field model, the container truck will dynamically adjust according to the speed and position of the vehicles in front and behind to ensure safe driving. Assume that the position of the container truck on the road is (x car ,y car ), the position information of the vehicle in front is (x front, y front ), the position information of the rear vehicle is (x rear ,y rear ), the relative speed between the truck and the surrounding vehicles is v r1 v r2 .

[0166] Accused vehicle v car =10m / s, the speed of the front and rear vehicles is v front =8m / s,v rear =12m / s. The longitudinal distance between the controlled vehicle and the preceding vehicle is K1 = 20m, the longitudinal distance between the controlled vehicle and the following vehicle is K2 = 15m, the lateral distance between the preceding vehicle and the controlled vehicle is M1 = 1.5m, the lateral distance between the following vehicle and the controlled vehicle is M2 = 0, the vehicle reaction time is ΔT = 1.5s, and the vehicle safety distance is s min =6m.

[0167] like Figure 8 As shown in the example diagram, according to the formula of the present invention, the longitudinal potential field of the controlled vehicle is as follows:

[0168] S=v r ·ΔT+s min =2*1.5+6=9m

[0169]

[0170] The environmental vehicle potential field is:

[0171]

[0172] The vehicle's lane line potential field is:

[0173]

[0174] The entire road potential field r mn for:

[0175]

[0176] The driving direction potential field is modeled as:

[0177]

[0178] The vehicle driving environment potential field is:

[0179] u mn =r mn +e mn +s mn

[0180] Generate a model predictive controller for the free cruising strategy of container trucks.

[0181] A model predictive controller (MPC) designed based on an environmental potential field and vehicle dynamics model can achieve safe, efficient, and smooth driving for autonomous vehicles. First, the controller optimizes the vehicle's future control inputs by calculating an objective function. The objective function consists of the environmental potential field value, velocity error, and control input increments. Higher potential field values ​​indicate greater risk. Control input increments are used to smooth changes in acceleration or steering angle, improving ride comfort and reducing energy consumption. The goal is to minimize driving risks, speed deviations, and sudden changes in control inputs. Second, the vehicle dynamics model predicts the vehicle's state over a period of time. Constraints on velocity, acceleration, and steering angle are incorporated to ensure that control inputs conform to vehicle physical characteristics and safety requirements. Through a rolling optimization approach, the MPC dynamically adjusts the control strategy based on the vehicle's current state and the real-time updates of the environmental potential field. Ultimately, the controller implements functions such as path tracking, speed regulation, and dynamic obstacle avoidance, effectively improving the adaptability and safety of autonomous vehicles in complex scenarios.

[0182] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A commercial vehicle free cruise control method, characterized by: include, Establish a potential field model of the truck driving environment; Design a predictive controller based on the free cruising strategy model of container trucks; Model predictive control specifies the objective function and solves the control input to perform target control; Generate predictive control trajectories to interact with distributed controllers of surrounding intelligent vehicles.

2. The commercial vehicle free cruise control method according to claim 1, characterized in that: The establishment of the potential field model of the container truck driving environment includes: Environmental potential field modeling and driving direction potential field modeling; The environmental potential field modeling includes obstacle vehicle potential field modeling and road potential field modeling; The driving direction potential field modeling includes providing a vehicle driving tendency toward the front of the lane, and changing lanes when an obstacle appears in front of the lane.

3. The commercial vehicle free cruise control method according to claim 1 or 2, characterized in that: The obstacle vehicle potential field modeling includes: Taking the longitudinal impact of the vehicle as the framework, the distribution of vehicle impact is described laterally; A local coordinate system is established with the vehicle body direction as the positive direction of the x-axis and the rear end point of the vehicle outer contour as the origin; Calculate the longitudinal potential field value of the environmental vehicle in sections; Determine the distance between the vehicle and the surrounding environment based on the relative speed; The longitudinal potential field risk distribution is reflected according to the vehicle distance relative to the environment.

4. The commercial vehicle free cruise control method according to claim 3, characterized in that: The obstacle vehicle potential field modeling also includes: The lateral potential field of the surrounding vehicle is calculated using a Gaussian-like function; The total value of the environmental vehicle potential field is obtained by combining the longitudinal potential field value of the environmental vehicle with the transverse potential field; Determine the location of the potential field calculation point around the vehicle and determine the lateral impact range of the potential field.

5. The commercial vehicle free cruise control method according to claim 2, characterized in that: The road potential field modeling includes: Establish lane line constraints through the road potential field, including whether vehicles can cross lane lines or not; When the lane line constraint relationship is that the lane line can be crossed, the potential field calculation point is set to calculate the generated road potential field; When the lane line constraint relationship is that the vehicle cannot cross the lane line, a quadratic curve is used to describe the influence of the road boundary on the road potential field, and the lane line constraint relationship is superimposed to obtain the entire road potential field.

6. The commercial vehicle free cruise control method according to claim 1, 2, 4 or 5, characterized in that: The designing of a predictive controller comprises, Optimize the vehicle's future control inputs by calculating the objective function through a distributed model predictive controller; In a fixed ground coordinate system, a predictive controller is designed by establishing a kinematic model of the container truck. The vehicle dynamics model predicts the vehicle state over a period of time in the future and controls the inputs based on the constraints; The vehicle kinematic model is optimized by using the constraint of increasing the control increment; Through rolling optimization, MPC dynamically adjusts the control strategy based on the vehicle's current state and the real-time updated environmental potential field to eliminate control blind spots.

7. The commercial vehicle free cruise control method according to claim 1, 2, 4 or 5, characterized in that: The objective function includes: The following vehicle control combines the environmental vehicle potential field, target speed, and economic comfort requirements, and combines the vehicle kinematic model with the calculation function of the potential field model to predict the potential field value of the driving environment potential field with the step length; Predict the lateral and longitudinal offset errors between the predicted trajectory of the following vehicle and the preceding vehicle in the time domain, predict the deviation between the speed and the target speed in the time domain, provide a reference target speed, and achieve the minimum incremental value considering the requirements of economy and comfort; Combined with the predicted control trajectory transmitted by the pilot vehicle, the objective function is constructed for tracking and the corresponding constraints are set.

8. The commercial vehicle free cruise control method according to claim 1, 2, 4 or 5, characterized in that: The target control comprises: The added linear constraints include vehicle speed control constraints and wheel deflection angle increment constraints.

9. A commercial vehicle free cruise control system, characterized in that: include, An environmental potential field modeling module (100) is used to establish a potential field model of the road and the surrounding vehicle, describe the interaction with the controlled vehicle's surrounding environment, and provide a basis for safety distance and risk assessment; A vehicle kinematics model module (200) is used to generate kinematic equations for the controlled vehicle, define the relationship between vehicle states and inputs, and support dynamic simulation of the control system; A model predictive control module (300) is used to design a controller to optimize the vehicle's travel along a target trajectory based on a vehicle kinematic model; An objective function design module (400) is used to construct an optimization objective function, including constraints on the accuracy of tracking the trajectory of the vehicle ahead, speed deviation, and control increment, to ensure that the optimization result meets driving requirements; The collaborative control decision module (500) is used to integrate information from various modules to perform multi-vehicle collaborative control to ensure the safety and stability of platoon driving.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the commercial vehicle free cruise control method are implemented.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the commercial vehicle free cruise control method are implemented.

Images