Intelligent automatic driving control system of bicycle

Abstract

The invention relates to the field of automatic driving, and discloses a bicycle intelligent automatic driving control system, which is configured to control each execution mechanism to complete actions through an action decision control module of the bicycle intelligent automatic driving control system, and the action decision control module comprises a map center line processing unit configured to obtain an initial map center line and generate a local planning track; a lateral control unit configured to construct and use a lateral predictive control model to output a lateral control signal; the longitudinal control unit is configured to output a speed predicted value by using a speed model established based on the acceleration response of the single vehicle under a preset condition; based on the deviation between the speed predicted value and the current actual state of the bicycle, estimating the current disturbing force of the bicycle; and calculating an expected accelerator force and an expected brake force for speed control according to the speed prediction value and the estimated current disturbance force. The problem that the steering wheel shakes due to the fact that the track is not smooth can be avoided, and the running stability and riding comfort of the vehicle are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving, and particularly to an autonomous driving control system for single-vehicle intelligence. Background Art

[0002] With the rapid development of artificial intelligence and related technologies, intelligent autonomous driving systems have become the core research direction of the automotive industry. A typical autonomous driving system usually includes key components such as an environmental perception subsystem, a positioning subsystem, a path planning subsystem, and a control decision-making subsystem.

[0003] In high-level autonomous driving systems, a high-precision positioning subsystem is the basis for safe and reliable operation. At present, although common positioning subsystems can integrate GPS / GNSS (Global Navigation Satellite System) and IMU (Inertial Measurement Unit) sensors with RTK (Real-Time Kinematic) technology to provide centimeter-level positioning, in scenarios where satellite signals are blocked (such as urban canyons, tunnels, indoor parking lots) or there is interference, the accuracy of such positioning subsystems will seriously decline or even fail. Some systems attempt to use SLAM (Simultaneous Localization and Mapping) modules based on lidar or vision sensors as a supplement or alternative. However, in the case of a lack of significant environmental features or drastic changes in dynamic environments, the robustness and long-term consistency of their outputs still face challenges. Therefore, existing positioning subsystems that rely solely on specific sensors or technologies are difficult to meet the stringent requirements of L4-level autonomous driving for high precision and high reliability in all scenarios.

[0004] The path planning subsystem of an autonomous driving system needs to generate safe, smooth, and traffic-rule-compliant driving trajectories in a complex and dynamic environment. Currently, some planning subsystems use traditional local path planning techniques, such as simple geometric curve fitting (e.g., cubic polynomials or spline curves). Although these techniques have a relatively low computational burden, they have inherent deficiencies in ensuring the smoothness of the generated trajectories, especially the continuity of curvature and curvature change rate. This may cause the vehicle control system to experience steering wheel jitter, poor ride comfort, or even unstable control during maneuvers such as high-speed driving or lane changes due to poor trajectory quality. Especially in scenarios where the attitude and curvature constraints of the trajectory start and end points need to be precisely met (e.g., requiring the start and end points to be in a straight-line driving state, i.e., zero curvature), the simple fitting techniques used in existing path planning subsystems are difficult to generate high-quality local trajectories.

[0005] One of the core tasks of the vehicle control subsystem is to accurately track the trajectory generated by the planning subsystem. In existing control subsystems, the design using traditional controllers (such as PID controllers, pure tracking algorithms, Stanley algorithms, etc.) has limited capabilities in dealing with the inherent nonlinear and time-varying characteristics of the vehicle, as well as coping with multiple constraints (such as the physical limitations of actuators, tire side slip limitations, and ride comfort requirements). Especially in high-speed or large-curvature working conditions, such control subsystems may have difficulty in balancing tracking accuracy and driving stability. Although the control subsystem based on model predictive control (MPC) is theoretically more suitable for handling such complex problems, its actual application effect is limited by factors such as the accuracy of its internal prediction model, the high requirements for computing resources, and how to effectively integrate vehicle stability control strategies. Therefore, there are still challenges in ensuring high-precision trajectory tracking and driving stability simultaneously in complex working conditions for existing control subsystems.

[0006] In terms of longitudinal control, many longitudinal control subsystems adopt PID-based speed control strategies. Such strategies are sensitive to changes in vehicle model parameters (such as changes in vehicle mass due to load changes) and disturbances in the external operating environment (such as slope resistance, air resistance), making it difficult for the system to achieve precise and robust speed tracking. It may show speed fluctuations, overshoots, or slow responses, thus affecting ride comfort and traffic flow efficiency. At the same time, how to enable the longitudinal control subsystem to intelligently adjust the adaptive cruise speed according to traffic rules, speed limit information, and map data is also the key to improving the user experience of the autonomous driving system. There is still room for improvement in the accuracy, robustness, and intelligent adaptive adjustment of speed control in existing longitudinal control subsystems. Summary of the Invention

[0007] To solve the deficiencies existing in the prior art, the present invention provides the following technical solutions: A single-vehicle intelligent autonomous driving control system, the system is configured to perform path planning according to single-vehicle positioning information, the current state of the single vehicle, and surrounding environment perception information, and through its action decision and control module, control each actuator to complete actions to achieve single-vehicle autonomous driving. The action decision and control module includes: A map centerline processing unit, configured to obtain an initial map centerline, and perform smooth fitting on the initial map centerline through a piecewise quintic polynomial to generate a local planning trajectory; A lateral control unit, configured to construct and utilize a lateral predictive control model, and take the local planning trajectory and the current state of the single vehicle as inputs, and output a lateral control signal; The longitudinal control unit is configured to utilize a speed model established based on the acceleration response of a single vehicle under preset conditions, take the current state of the single vehicle as an input, and output a speed prediction value; estimate the interference force currently acting on the single vehicle based on the deviation between the speed prediction value and the current actual state of the single vehicle; calculate the desired throttle and desired braking force for speed control according to the speed prediction value and the estimated current interference force.

[0008] Preferably, when the map centerline processing unit performs smooth fitting on the initial map centerline through a piecewise quintic polynomial, the set constraint conditions include: the curvature of the local planning trajectory at the initial position and the target position of each trajectory segment formed by the fitting is zero.

[0009] Preferably, the map centerline processing unit is configured to generate the local planning trajectory through the following steps: S101. For the currently to-be-generated trajectory segment, set its position and heading angle at the initial position and the target position of this segment as further constraint conditions. S102. Based on the position at the initial position and the target position of this segment set in step S101, the heading angle, and the constraint condition that the curvature at the initial position and the target position of this segment is zero, solve the coefficients of the piecewise quintic polynomial for generating this current trajectory segment. S103. Generate this current trajectory segment according to the coefficients solved in step S102 as a part of the local planning trajectory.

[0010] Preferably, the lateral control unit is specifically configured to output a lateral control signal in the following manner: S201. Establish a prediction model based on the kinematics of the vehicle model. S202. Define a cost function within the prediction horizon, where the cost function aims to minimize the deviation between the predicted state and the ideal state, the magnitude of the control input or the deviation from the ideal control input, and the deviation between the predicted terminal state and the ideal terminal state. S203. Set the constraint conditions for the optimization solution, where the constraint conditions include: the kinematic constraints of the vehicle model, the constraint with the current state of the single vehicle as the initial state, and the control input quantity constraint. S204. By solving the optimization problem in each control cycle to minimize the cost function on the premise of satisfying the constraint conditions, obtain the optimal control input sequence, and generate the lateral control signal according to this optimal control input sequence.

[0011] Preferably, the lateral control unit is further configured to output a lateral control signal in the following manner: S211. Estimate the current centroid side slip angle of the single vehicle based on a preset two-degree-of-freedom vehicle model and a state observer; S212. Conduct a vehicle yaw stability analysis to determine the yaw stability condition in the current state; S213. Calculate an additional yaw moment for satisfying the yaw stability condition, and adjust the lateral control signal according to the additional yaw moment.

[0012] Preferably, the state variables used in the lateral predictive control model include: lateral error, change rate of lateral error, heading error, and change rate of heading error.

[0013] Preferably, the map centerline processing unit is further configured to optimize the local planned trajectory, which is achieved through the following steps: S111. Generate multiple alternative local trajectories covering a predetermined time range; S112. Set an evaluation function for evaluating the alternative local trajectories, and the evaluation function at least includes a cost term for evaluating the lateral offset of the single vehicle from the map centerline and a cost term for evaluating the comfort of the trajectory; S113. Calculate the evaluation function value corresponding to each alternative local trajectory; S114. Select the alternative local trajectory with the optimal evaluation function value as the planned trajectory.

[0014] Preferably, the longitudinal control unit obtains the current disturbing force in the following ways: S301. Calculate the speed residual between the actually measured speed of the single vehicle and the speed prediction value; S302. Generate an estimated value of the disturbing force by performing proportional-integral processing on the speed residual.

[0015] Beneficial effects In the present invention, the map centerline processing unit uses a piecewise quintic polynomial to perform smooth fitting on the map centerline, and can use the curvatures of the starting point and the ending point (for example, set to zero) as constraint conditions, so as to be able to generate a local planned trajectory with more continuous and smoother curvature and curvature change rate. This configuration of the system effectively avoids the problem of steering wheel jitter caused by unsmooth trajectories, especially in scenarios such as high-speed driving or when precise control of the starting and ending states (such as straight-line merging or leaving) is required, significantly improving the driving smoothness and riding comfort of the vehicle.

[0016] By using the model predictive control (MPC) strategy in its lateral control unit, the present invention can prospectively consider vehicle dynamics and trajectory deviations over a period of time in the future, systematically optimize control inputs (such as steering angles), and thus achieve higher-precision tracking of the planned trajectory. At the same time, this lateral control unit can explicitly handle the kinematic / dynamic constraints of the vehicle and control quantity constraints to ensure the safety and feasibility of control commands. The system can also integrate a yaw stability correction function based on a two-degree-of-freedom model and a state observer, enabling it to actively estimate and compensate for potential instability risks, and further enhancing the driving stability of the vehicle under dynamic limits or low-adhesion road surfaces.

[0017] The longitudinal control unit of the present invention adopts a speed model established based on vehicle acceleration response and combines the analysis of real-time speed deviations to estimate the total disturbance force currently received (including unmodeled factors such as slopes, loads, and wind resistance), enabling this unit to prospectively calculate the desired driving / braking force according to the predicted speed and the estimated disturbance force. Compared with a system using traditional PID control, this design of the present system can more effectively suppress the influence of external disturbances on speed, improve the accuracy and robustness of speed tracking, reduce unnecessary acceleration and deceleration, enhance ride comfort and potential fuel / energy economy, and better adapt to changes in the vehicle's own state (such as load). Brief Description of the Drawings

[0018] Figure 1 It is a schematic structural diagram of a single-vehicle intelligent autonomous driving control system provided in a preferred embodiment of the present invention; Figure 2 It is a schematic diagram comparing the trajectory effects generated by a cubic polynomial adopted by the prior art in the field and a fifth-degree polynomial adopted by the present invention. Detailed Embodiment

[0019] Embodiment 1 As Figure 1 shown, the present invention provides a single-vehicle intelligent autonomous driving control system. The system is configured to perform path planning based on single-vehicle positioning information, the current state of the single vehicle, and surrounding environment perception information, and control each actuator to complete actions through its action decision and control module to achieve single-vehicle autonomous driving.

[0020] Among them, the action decision control module refers to a key processing unit within the system. After higher-level planning (such as completed by the global path planning module or behavior decision module to determine the general driving route or intention), it is responsible for converting the planned driving intention or reference path (such as the map center line) into specific and executable vehicle low-level control instructions. This action decision control module is a key bridge connecting "macro path planning" and "micro vehicle execution" within the system. Its function is to convert the abstract planning intention of "where to go and how to go" into precise and real-time control commands (Commands) sent to the vehicle low-level actuators (steering, driving, braking systems) through the specific trajectory generation, lateral control, and longitudinal control algorithms integrated within it, so as to directly drive the vehicle to complete specific driving actions (such as lane keeping, turning, accelerating and decelerating, etc.).

[0021] The single-vehicle intelligent autonomous driving control system provided by the present invention, its action decision control module includes: A map center line processing unit, which is configured to obtain the initial map center line and perform smooth fitting on the initial map center line through a piecewise quintic polynomial to generate a local planning trajectory.

[0022] It should be understood that the initial map center line input to this map center line processing unit represents the geometric center of the lane or path that the vehicle expects to drive on. This initial center line may directly come from the high-precision map database docked by the system, or be generated by a higher-level path planning module within the system (such as a task-based or behavior-based planner) according to navigation tasks and real-time environment perception information. Its form may be a discrete point sequence or a rough curve. However, such raw input data often has inherent problems. For example, it may consist of discrete points, contain noise, or be geometrically not smooth enough, especially there may be discontinuities or even mutations in curvature. If the system directly performs vehicle tracking control based on such raw center lines, especially at higher speeds, it will cause the output control commands (especially steering commands) to be adjusted frequently and violently, which may cause the steering wheel to shake, reduce ride comfort, and even affect the driving stability under the system control. Therefore, it is necessary for this map center line processing unit to perform smooth fitting processing on it.

[0023] In some existing autonomous driving systems, their trajectory planning modules typically use cubic polynomials or cubic B-spline functions to generate a cluster of alternative trajectories. These modules take the current position and heading angle of the vehicle as the initial state constraints, and the positions and heading angles of the sampled target points on the road as the target state constraints, and accordingly generate a curved trajectory connecting the initial state and the target state. However, the systems adopting such technologies can only ensure the first-order continuity of the steering angle corresponding to the generated trajectory, and cannot guarantee the continuity of the steering angular velocity (i.e., the second-order continuity). This limitation in the system design has relatively little impact under low-speed driving conditions, and the discontinuity can be ignored; but under high-speed driving conditions, such discontinuity will be significantly amplified, which may cause unstable oscillations of the steering system under system control. In addition, the planning module based on cubic polynomials has inherent limitations in dealing with specific boundary conditions. As Figure 2 shown, when the curvatures of the initial state and the target state of the vehicle are both zero, such modules cannot simultaneously satisfy the zero-curvature constraint conditions at these two places (shown by the dotted line in the figure), resulting in a curvature mutation during the local trajectory update process and causing unstable steering control of the system.

[0024] In view of this, the map centerline processing unit of the present invention innovatively uses the quintic polynomial method to generate the local planning trajectory. The quintic polynomial has sufficient mathematical degrees of freedom, enabling this unit to simultaneously satisfy the continuity constraints of position, heading angle, and curvature. During the actual fitting process, this unit processes the initial map centerline in segments, and introduces the curvature values of the initial state and the target state as additional constraint conditions into the polynomial solution process, applies the quintic polynomial function to each segment of the path for precise fitting, and at the same time ensures a high-order smooth transition at the connection of the segments.

[0025] Specifically, the quintic polynomial is a function in the following form: ; Among them, x(t) is the horizontal coordinate at time t; y(t) is the vertical coordinate at time t. The horizontal and vertical positions of each quintic polynomial segment are determined by 6 coefficients (a0 to a5, b0 to b5) respectively. This representation provides the map centerline processing unit with sufficient degrees of freedom (6 coefficients per segment), enabling it to simultaneously satisfy the three key geometric constraints of each path segment at the starting and ending points, namely the position, tangent direction (corresponding to the heading angle), and curvature, and even the rate of change of curvature can be constrained. This unit determines the polynomial coefficients of each segment by solving these constraint conditions, and finally stitches these segments together to generate a locally planned trajectory that is globally smooth (at least curvature continuous, i.e., C2 continuity). The trajectory generated by this unit not only closely adheres to the intention of the original centerline, but more importantly, it ensures a high degree of smoothness in the internal geometric properties of the trajectory, providing a high-quality tracking target for the subsequent lateral and longitudinal control units of the system, thus ensuring that the vehicle can perform the autonomous driving task smoothly, precisely, and comfortably under the system control.

[0026] In the map centerline processing unit of this system, the entire path is not represented by a single quintic polynomial function. Instead, this unit cuts the path into multiple continuous segments for processing. Each path segment is described by its own independent set of quintic polynomials. This unit ensures that these segments are smoothly connected at the connection points (referred to as nodes or knots). The advantage of using segmented processing is that relatively low-order polynomials (quintics are chosen to satisfy sufficient constraint conditions) can be used to fit complex paths, avoiding the Runge phenomenon that may occur with a single high-order polynomial (i.e., severe oscillations at the endpoints). At the same time, the calculation of each segment is relatively independent, facilitating the processing of this unit.

[0027] The Smooth Fitting operation performed by the map centerline processing unit refers to given a set of discrete data points or a possibly non-smooth initial curve as input, this unit searches for and generates a new, smooth curve. This new curve is as close as possible to the original input data or curve under a certain metric (such as distance, shape), and itself satisfies certain smoothness requirements. Its purpose is to generate a curve that is easy to analyze, calculate, and easy for the vehicle to accurately and comfortably track in autonomous driving applications. In the prior art, some systems may use spline fitting and other methods to complete similar tasks.

[0028] In many actual driving scenarios, a vehicle needs to transition between straight-line driving and curved driving. For example, when entering or exiting a curve, when returning to the center of the lane in a straight line after completing a lane change, or when starting from a standstill or preparing to stop, it is usually expected that the vehicle's trajectory is smooth and natural at the transition points. If the curve segment generated by the system planning unit still has non-zero curvature at the starting point or the ending point, then there will be a sudden change in curvature when connecting to the straight-line segment. This sudden change will require the steering control unit of the system to output an instruction to instantaneously change the steering angle, which may cause jerking or shaking of the steering wheel, affecting ride comfort and control stability.

[0029] Therefore, in some preferred embodiments, the map centerline processing unit is configured such that: when performing smooth fitting on the initial map centerline through a piecewise quintic polynomial, the set constraint conditions include that the curvature of the local planned trajectory is zero at the initial position and the target position of each trajectory segment formed by the fitting. By imposing the above constraints, the unit can achieve smooth and continuous transition between the generated trajectory and the straight-line segment, thereby improving the smoothness, comfort, and control stability of vehicle driving, and enabling the system to better simulate natural driving behavior and meet the requirements of specific driving tasks. This demonstrates the ability of the map centerline processing unit of the present system to effectively utilize the high-degree-of-freedom characteristics of the quintic polynomial to optimize the trajectory quality. As Figure 2 shown by the black solid line in, the trajectory generated by the unit not only ensures continuity in the position and heading angle dimensions, but more importantly, achieves continuity of the curvature and its derivative (curvature change rate), enabling the vehicle to achieve high-precision smooth steering when the system performs steering control, effectively eliminating sudden changes and oscillations in the steering system, and significantly enhancing the control stability and driving safety of the vehicle, especially under high-speed working conditions.

[0030] In some preferred embodiments, the map centerline processing unit of the present system is configured to precisely construct each component of the local planned trajectory through the following steps: including: S101. For the currently to-be-generated trajectory segment, set the position and heading angle at the initial position and the target position of this segment as further constraint conditions. That is, set the precise position coordinates (x, y) and heading angle at the initial time (t = 0) and the target time (t = T) of this trajectory segment. These values are obtained from the sampling analysis of the initial map centerline, and are used to ensure seamless docking with the end state (position and heading) of the previously generated trajectory segment, and point to the expected starting state of the next trajectory segment. Specifically, the above constraint conditions can be expressed by the following formulas in the x and y directions: ; ; ; ; ; wherein, is the horizontal coordinate at time t; is the vertical coordinate at time t; is the speed of the bicycle in the x - direction at time t, is the speed of the bicycle in the y - direction at time t; is the acceleration of the bicycle in the x - direction at time t; theta is the heading angle, k is the curvature, and v is the speed.

[0031] S102. Based on the positions at the initial position and the target position of this segment set in step S101, the heading angle, and the constraint condition that the curvature is zero at the initial position and the target position of this segment, solve for the coefficients of the piecewise quintic polynomial used to generate the current trajectory segment. The above six constraint conditions jointly describe the six undetermined coefficients contained in each of the two quintic polynomials required for the x - coordinate and y - coordinate of this trajectory segment. Based on this, a linear equation system can be constructed for solution. The linear equation system is expressed in matrix form as follows: ; wherein, , , respectively represent the horizontal coordinate, speed, and acceleration at the initial position; , , respectively represent the horizontal coordinate, speed, and acceleration at the position where time t1 is located; , , respectively represent the vertical coordinate, speed, and acceleration at the initial position; , , respectively represent the vertical coordinate, speed, and acceleration at the position where time t1 is located; It should be understood that the times t0 and t1 are known and represent the times of the initial position and the end position respectively. Therefore, the X, Y, and T matrices are all known, and the coefficient matrices A in the x - direction and B in the y - direction can also be obtained.

[0032] S103. Generate the current trajectory segment based on the coefficients obtained in step S102 as a part of the local planned trajectory. By continuously varying the time parameter t from 0 to T and substituting it into the two fifth-degree polynomial equations x(t) and y(t), the precise vehicle position coordinates corresponding to any moment within this time period are calculated. Connecting these consecutive position points generates the current smooth trajectory curve. This curve segment is then regarded as an integral part of the entire local planned trajectory, stored, and used as the starting point for the calculation of the next trajectory segment (providing the initial position, heading angle, and curvature constraints for the next segment). By repeatedly executing the three steps S101 to S103, segment by segment generation and splicing are performed, ultimately forming a complete local planned trajectory that meets all constraint conditions and is highly smooth for the vehicle control system to track.

[0033] Furthermore, if the time interval t cannot be directly given at the moments of t0 and t1, assume the constant speed of the single vehicle is v to estimate the time from point A to point B: . When the independent variable of the polynomial curve is time t, and the coefficient matrices A and B have been obtained previously, the position, speed, etc. of each point on the curve are then determined, and the trajectory can be obtained. Specifically, the calculation methods for the heading angle theta, curvature kappa, and curvature change rate dkappa at any point on the local planned trajectory are as follows: ; ; ; Among them, is the lateral coordinate at time t; is the longitudinal coordinate at time t; is the speed of the single vehicle in the x direction at time t, is the speed of the single vehicle in the y direction at time t; is the acceleration of the single vehicle in the x direction at time t; is the acceleration of the single vehicle in the y direction at time t.

[0034] In some other preferred embodiments, the single smooth trajectory generated by the foregoing process of the system may be merely the one that is geometrically the smoothest or closest to the center line. However, to enable the system to find the driving path that optimally performs in multiple evaluation dimensions and achieve a decision-making that pursues higher comfort, efficiency, and intelligence on the basis of meeting basic safety and driving requirements, the action decision control module of the system (or the map center line processing unit inside it) can also be configured to execute an optimization function for the local planned trajectory. This optimization function is implemented by this unit / module through the following steps: S111. Generate multiple alternative local trajectories that cover a predetermined time range. It should be understood that there are many ways to generate multiple slightly different but all potentially safe and feasible path options as alternative local trajectories. For example, it can be done by making small perturbations to the base trajectory, or by using different parameters (such as different target speeds, different avoidance strategy parameters) to drive the solution of the quintic polynomial to produce different results. The present invention does not make further limitations in this regard.

[0035] S112. Set an evaluation function for evaluating the alternative local trajectories. This evaluation function at least includes a cost term for evaluating the lateral offset of the single vehicle from the center line of the map, and a cost term for evaluating the comfort of the trajectory.

[0036] The cost term for evaluating the lateral offset of the single vehicle from the center line of the map is used to penalize those trajectories that deviate too far from the desired driving path, so as to ensure that the vehicle stays within the lane as much as possible or follows a predetermined route.

[0037] The cost term for evaluating the comfort of the trajectory is used to penalize those trajectories that include sharp accelerations and decelerations, sharp turns, or high-frequency oscillations. Usually, it is achieved by calculating the jerk or the magnitude of the curvature and the rate of change of curvature of the trajectory, with the aim of ensuring a smooth and comfortable ride for passengers.

[0038] In some other preferred embodiments, the evaluation function may further include other important factors, such as the collision risk with perceived static or dynamic obstacles (the closer the distance, the higher the cost), the total length or travel time of the trajectory (evaluating efficiency), energy consumption, the deviation from the target speed, etc.

[0039] S113. Calculate the evaluation function value corresponding to each alternative local trajectory. This means that along the time or space process of each alternative trajectory, all relevant cost terms (multiplied by their respective weights) are accumulated (or integrated), and finally a total evaluation value or total cost score of this trajectory is obtained.

[0040] S114. Select the alternative local trajectory with the optimal evaluation function value as the planned trajectory. Obviously, the planned trajectory is the best solution obtained by comprehensively considering path keeping, ride comfort, and possible other factors (such as safety, efficiency) under the current evaluation criteria and weight system. This optimization selection process enables autonomous driving to not only mechanically follow a line, but also make intelligent decisions in the possibility space to achieve better overall driving performance.

[0041] The lateral control unit is configured to construct and utilize a lateral predictive control model, which takes the local planned trajectory and the current state of the vehicle as inputs and outputs a lateral control signal. It should be understood that the lateral control undertakes the core vehicle tracking function, and its fundamental goal is to accurately and smoothly control the steering system of the vehicle so that the vehicle can closely follow the local planned trajectory.

[0042] The lateral predictive control model (Lateral Model Predictive Control, MPC) can be implemented by establishing a mathematical model that can describe the lateral motion behavior of the vehicle. Specifically, a discrete state-space form can be used to capture how key state variables such as the lateral position, orientation angle (yaw angle), lateral velocity, and yaw angular velocity of the vehicle respond to changes in the steering input (control variable, usually the front wheel angle or the steering wheel angle). It should be understood that the establishment of the model is based on simplified vehicle kinematics or more complex dynamics principles. Since this part of the content is not the focus of the present invention, it will not be elaborated here. Those skilled in the art can make specific designs according to the existing technology. In some other preferred embodiments, the lateral control unit is specifically configured to output the lateral control signal in the following manner, which specifically includes: S201. Establish a prediction model based on vehicle model kinematics.

[0043] S202. Define the cost function within the prediction time domain. The cost function aims to minimize the deviation between the predicted state and the ideal state, the magnitude of the control input or its deviation from the ideal control input, and the deviation between the predicted terminal state and the ideal terminal state.

[0044] The design of the cost function aims to minimize the costs in multiple aspects simultaneously: one is the deviation between the predicted state and the ideal state (i.e., the target state on the planned trajectory), which ensures that the vehicle can closely track the planned path, reducing the lateral error and the orientation error; the second is the magnitude of the control input or its deviation from the ideal control input, which encourages the controller to use smoother and more economical steering actions, avoiding overly aggressive operations, improving the riding comfort and reducing actuator wear; the third is the deviation between the predicted terminal state and the ideal terminal state, which helps to ensure that the vehicle is in a good and stable state at the end of the prediction time domain, preparing for the subsequent control cycle and enhancing the long-term stability of the system. By assigning reasonable weights to these different cost terms, trade-offs and adjustments can be made among multiple objectives such as tracking accuracy, control smoothness, and stability to meet the requirements of different driving scenarios.

[0045] S203. Set the constraint conditions for the optimal solution. The constraint conditions include: the kinematic constraints of the vehicle model, the constraints with the current state of the single vehicle as the initial state, and the control input quantity constraints. The constraint conditions define the boundaries and rules that must be followed during the optimal solution process, specifically including: 1. The kinematic / dynamic constraints of the vehicle model, that is, any predicted evolution of the vehicle state must strictly follow the physical laws described by the mathematical model established in S201; 2. The constraints with the current state of the single vehicle as the initial state, which means that all predictions and optimizations must be based on the actual measured or estimated state of the vehicle at the start of the current control cycle (such as the current position error, heading error, etc.), ensuring that the control decisions are based on the actual situation; 3. Third, the control input quantity constraints, which represent the limitations of the vehicle's physical actuators, such as the maximum steering angle limit of the steering wheel or the front wheels, the maximum steering angular velocity limit, etc., to ensure that the calculated steering command does not exceed the capabilities of the vehicle hardware.

[0046] It should be understood that there may also be other constraints, such as the maximum allowable tire side slip angle, the maximum allowable lateral acceleration (for comfort or safety), etc. Optimizing within the feasible region formed by these constraints is the key to ensuring the effectiveness and safety of the MPC output result.

[0047] S204. By solving the optimization problem in each control cycle, to minimize the cost function under the premise of satisfying the constraint conditions, obtain the optimal control input sequence, and generate the lateral control signal according to this optimal control input sequence. It should be understood that the optimal control input sequence covers the entire prediction time domain. The lateral prediction control model will extract one of the control inputs in this sequence (it can be the first one, or other control inputs can be selected according to other preset rules), and the remaining control inputs in the sequence are discarded. In the next control cycle, repeat the above steps. This way of continuously re-planning and correcting the control strategy based on the latest information makes the MPC have strong robustness and can effectively cope with model uncertainties, external disturbances, and environmental changes, so as to achieve continuous, accurate, and stable lateral control.

[0048] Those skilled in the art can know that when the vehicle is approaching or in a potentially unstable state (such as high-speed cornering, emergency obstacle avoidance, or driving on a low-adhesion road surface), it is necessary to actively intervene and adjust the control to maintain or restore the driving stability of the vehicle and prevent situations such as sideslip, fishtailing, or even loss of control. In some preferred embodiments, the above lateral control unit is further configured to output the lateral control signal in the following manner, specifically including: S211. Estimate the current centroid side slip angle of the vehicle based on a preset two-degree-of-freedom vehicle model and a state observer. Those skilled in the art can understand that the centroid side slip angle refers to the angle between the velocity direction at the vehicle's centroid and the vehicle's longitudinal axis, and it is the most direct indicator to measure the degree of vehicle side slip. Since there is no sensor that can directly measure this angle, it is usually difficult to directly measure. This embodiment provides an indirect estimation based on a preset two-degree-of-freedom (2-DOF) vehicle model. Specifically, it can be a simplified vehicle dynamics model that mainly considers the lateral motion of the vehicle and the yaw motion around the vertical axis, which captures how the lateral velocity and yaw angular velocity of the vehicle respond to the steering input and tire lateral force. To use this model for estimation, it is also necessary to combine the actual measurement information of the vehicle (such as yaw angular velocity, lateral acceleration, wheel speed, steering angle, etc.), and estimate the current centroid side slip angle of the vehicle in real time by designing and running a state observer (such as a Kalman filter or its variants, such as the Extended Kalman Filter EKF, Unscented Kalman Filter UKF, or a sliding mode observer, etc.). This state observer can fuse the model prediction and sensor measurement values, overcome measurement noise and model uncertainty.

[0049] S212. Conduct a vehicle yaw stability analysis to determine the yaw stability condition in the current state. It should be understood that the yaw stability of the vehicle is usually closely related to two state variables, namely the centroid side slip angle and the yaw angular velocity. The stability analysis can be based on various theories or methods, such as phase plane analysis, linear or nonlinear stability theories (such as Lyapunov stability theory), etc., and the present invention does not make further limitations.

[0050] S213. Calculate the additional yaw moment for satisfying the yaw stability condition, and adjust the lateral control signal according to the additional yaw moment. The additional yaw moment is a virtual or actual additional rotational moment applied to the vehicle around the vehicle's vertical axis, and its function is to help the vehicle resist excessive steering or understeering tendencies and pull the vehicle state back into the stable region. For example, if the vehicle exhibits excessive steering (fishtailing), a reverse (rotation-inhibiting) yaw moment needs to be applied; if there is understeering (pushing), a yaw moment that helps the vehicle turn needs to be applied. Its magnitude and direction can be calculated according to the stability condition determined in step S212 and the distance between the current vehicle state and the stability boundary.

[0051] The longitudinal control unit is configured to use a speed model established based on the acceleration response of the vehicle under preset conditions, take the current state of the vehicle as input, and output a speed prediction value; estimate the interference force currently received by the vehicle based on the deviation between the speed prediction value and the current actual state of the vehicle; calculate the desired throttle and desired braking force for speed control according to the speed prediction value and the estimated current interference force.

[0052] The longitudinal control unit is responsible for managing the forward and backward movement of the vehicle. The core task is to precisely control the speed of the vehicle so that it can safely, smoothly and efficiently follow the planned speed instruction or achieve adaptive cruise control.

[0053] The speed model can be a parameterized mathematical model (such as a transfer function, a state-space equation), or a data-driven model (such as a lookup table, a neural network). This model takes the current state of the vehicle (including at least the currently measured speed, and possibly also the gear position, the current control input, etc.) as input, and then outputs a speed prediction value. This prediction value represents how the vehicle speed should change in the next very short time step under the current state, and it predicts the speed evolution without unmodeled disturbances.

[0054] Those skilled in the art can understand that the actual speed response of the vehicle often deviates from the prediction of the above ideal model because there are many factors that are not precisely included in the model or change in real time, such as changes in road slope, changes in real-time wind resistance (headwind / tailwind), changes in vehicle load, changes in rolling resistance (such as changes in road surface material), etc. These unmodeled factors together constitute a disturbing force, which affects the actual acceleration of the vehicle. In some preferred embodiments, the longitudinal control unit obtains the current disturbing force in the following ways: S301. Calculate the speed residual between the actually measured speed of the vehicle and the speed prediction value. This residual signal directly quantifies the mismatch between the model prediction and the actual situation. If the residual is zero or close to zero, it indicates that the model can well explain the current speed behavior; if the residual persists in being non-zero, it strongly implies that there is an external force (i.e., the disturbing force) not considered by the model that is affecting the movement of the vehicle - a positive residual may mean there is assistance (such as going downhill), and a negative residual may mean there is resistance (such as going uphill or against the wind). Therefore, the speed residual is the original input signal for detecting and quantifying the disturbing effect.

[0055] S302. Generate an estimated value of the disturbing force by performing proportional-integral processing on the speed residual. Those skilled in the art can understand that simply regarding the speed residual directly as the disturbing force is not accurate because it is only the result of the disturbing effect (speed change), rather than the disturbing force itself. This step converts the measurable speed deviation into a quantitative estimate of the non-directly measurable disturbing force, providing a key basis for subsequent longitudinal control compensation. Among them, proportional-integral (PI) processing is a conventional technique in the art, and the present invention will not be further described.

[0056] The calculation of the desired throttle opening (for acceleration or speed maintenance) and the desired braking force (for deceleration) for final speed control is simultaneously based on (or, considered in a feed-forward manner) the aforementioned speed prediction value (associated with the basic driving / braking force required to reach the target speed) and the currently estimated current disturbance force (as a compensation term for the basic driving / braking force). For example, if a large uphill resistance is estimated (the disturbance force is negative), the controller will increase the additional throttle opening on top of the basic throttle opening calculated by feed-forward to compensate for this resistance, so as to ensure that the vehicle can still reach the target acceleration or maintain the target speed. Conversely, if a downhill assist is estimated (the disturbance force is positive), the throttle may be reduced or even a slight brake may be required. This longitudinal control strategy that combines model prediction and real-time estimation and compensation of disturbance forces enables the system to have stronger robustness and adaptability to external environmental changes and vehicle own parameter changes (such as load), can achieve more accurate and smoother speed tracking, and improves the comfort and reliability of autonomous driving.

[0057] The above has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A bicycle intelligent automatic driving control system, which is configured to perform path planning based on bicycle positioning information, the current state of the bicycle, and surrounding environment perception information, and control each actuator to complete actions through its action decision-making control module to achieve automatic driving of the bicycle, characterized in that, The action decision and control module includes: A map centerline processing unit configured to obtain an initial map centerline, perform smooth fitting on the initial map centerline through piecewise quintic polynomials, and generate a local planning trajectory; A lateral control unit configured to construct and utilize a lateral predictive control model, take the local planning trajectory and the current state of the bicycle as inputs, and output a lateral control signal; A longitudinal control unit configured to utilize a speed model established based on the acceleration response of the bicycle under preset conditions, take the current state of the bicycle as an input, and output a speed prediction value; estimate the interference force currently acting on the bicycle based on the deviation between the speed prediction value and the current actual state of the bicycle; calculate the desired throttle and desired braking force for speed control according to the speed prediction value and the estimated current interference force.

2. The bicycle intelligent automatic driving control system according to claim 1, characterized in that: When the map centerline processing unit performs smooth fitting on the initial map centerline through piecewise quintic polynomials, the set constraint conditions include: the curvature of the local planning trajectory at the initial position and the target position of each trajectory segment formed by the fitting is zero.

3. The bicycle intelligent automatic driving control system according to claim 2, characterized in that, The map centerline processing unit is configured to generate the local planning trajectory through the following steps: S101. For the currently to-be-generated trajectory segment, set the position and heading angle at the initial position and the target position of this segment as further constraint conditions; S102. Based on the position at the initial position and the target position of this segment set in step S101, the heading angle, and the constraint condition that the curvature at the initial position and the target position of this segment is zero, solve the coefficients of the piecewise quintic polynomial for generating this current trajectory segment; S103. Generate this current trajectory segment according to the coefficients solved in step S102 as a part of the local planning trajectory.

4. The bicycle intelligent automatic driving control system according to claim 1, characterized in that, The lateral control unit is specifically configured to output a lateral control signal in the following manner: S201. Establish a prediction model based on the kinematics of the vehicle model; S202. Define a cost function within the prediction time domain, and the cost function aims to minimize the deviation between the predicted state and the ideal state, the magnitude of the control input quantity or the deviation from the ideal control input, and the deviation between the predicted end state and the ideal end state; S203. Set the constraint conditions for the optimization solution, and the constraint conditions include: the kinematic constraints of the vehicle model, the constraint with the current state of the bicycle as the initial state, and the control input quantity constraint; S204. By solving the optimization problem in each control cycle, minimize the cost function on the premise of satisfying the constraint conditions, obtain an optimal control input sequence, and generate the lateral control signal according to this optimal control input sequence.

5. The bicycle intelligent automatic driving control system according to claim 4, characterized in that, The lateral control unit is also configured to output a lateral control signal in the following manner: S211. Estimate the current sideslip angle of the center of mass of the bicycle based on a preset two-degree-of-freedom vehicle model and a state observer; S212. Conduct vehicle yaw stability analysis to determine the yaw stability condition under the current state; S213. Calculate an additional yaw moment for satisfying the yaw stability condition, and adjust the lateral control signal according to this additional yaw moment.

6. The bicycle intelligent automatic driving control system according to claim 4, characterized in that, The state variables used in the lateral prediction control model include: lateral error, the change rate of lateral error, heading error, and the change rate of heading error.

7. The bicycle intelligent automatic driving control system according to claim 1, characterized in that, The map centerline processing unit is further configured to optimize the local planned trajectory, which is achieved through the following steps: S111. Generate multiple alternative local trajectories covering a predetermined time range; S112. Set an evaluation function for evaluating the alternative local trajectories, where the evaluation function at least includes a cost term for evaluating the lateral offset of the single vehicle from the map centerline and a cost term for evaluating the comfort of the trajectory; S113. Calculate the evaluation function value corresponding to each alternative local trajectory; S114. Select the alternative local trajectory with the optimal evaluation function value as the planned trajectory.

8. The bicycle intelligent automatic driving control system according to claim 1, characterized in that, The manner in which the longitudinal control unit obtains the current disturbing force includes: S301. Calculate the speed residual between the actually measured speed of the single vehicle and the speed prediction value; S302. Generate an estimated value of the disturbing force by performing proportional-integral processing on the speed residual.

Images