CN-122009192-A - Unmanned vehicle road surface disturbance prediction and active stability control method

CN122009192ACN 122009192 ACN122009192 ACN 122009192ACN-122009192-A

Abstract

The invention belongs to the technical field of vehicle stability control, and particularly relates to an unmanned vehicle road surface disturbance prediction and active stability control method. The method comprises the following steps of S1, collecting road environment information and vehicle motion states, S2, enabling multi-source information to be time-synchronous and space-unified, S3, locally modeling three-dimensional road surface geometry, S4, predicting and modeling road surface disturbance, S5, calculating disturbance feedforward compensation control quantity, S6, calculating gesture feedback stability control, S7, fusing disturbance feedforward compensation control quantity and gesture feedback control quantity to obtain final stability control input of a vehicle, and S8, executing mechanism gesture adjustment, wherein the vehicle executing mechanism is controlled and adjusted according to the stability control input. The invention can improve the gesture stability and disturbance rejection capability of the unmanned vehicle under the condition of complex road surfaces, and enhance the running safety and adaptability of the vehicle in actual application scenes.

Inventors

YU HAIYE
PAN ZHIHAO
ZHANG LEI
SUI YUANYUAN
Fu Hanbing
WANG LINGSHUANG
ZHANG CHENXI

Assignees

吉林大学

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (9)

1. The unmanned vehicle road surface disturbance prediction and active stability control method is characterized by comprising the following steps of: s1, acquiring road environment information and vehicle motion state, namely acquiring road space information in front of a vehicle and vehicle motion state information in real time, and establishing a three-dimensional road surface set; s2, multi-source information time synchronization and space unification, namely uniformly processing the acquired multi-source data; S3, local three-dimensional road surface geometric modeling, namely, based on the synchronized road surface point cloud data, carrying out three-dimensional geometric modeling on a vehicle front area, and further calculating road surface geometric characteristic parameters including road surface gradient, road surface curvature and road surface roughness after obtaining a road surface model; S4, road disturbance prediction modeling, namely establishing a disturbance prediction model facing to vehicle dynamics response according to the current motion state of the vehicle and road geometry characteristics, and estimating disturbance to be received by the vehicle in a future time window; S5, calculating disturbance feedforward compensation control quantity according to the predicted future disturbance, so that the control system can perform pre-regulation before the disturbance action; s6, attitude feedback stable control calculation, namely establishing a vehicle attitude feedback control mechanism based on feedforward compensation, and specifically adopting a cascade control structure, wherein an attitude angle control loop generates expected angular velocity according to the error between a target attitude and an actual attitude; s7, fusing feedforward control and feedback control, namely fusing disturbance feedforward compensation control quantity and attitude feedback control quantity to obtain final stability control input of the vehicle; And S8, performing attitude adjustment on the actuating mechanism, namely performing control adjustment on the actuating mechanism of the vehicle according to the stability control input.
2. The unmanned vehicle road disturbance prediction and active balance control method according to claim 1, wherein in the step S1, road environment information comprises road surface image information and road surface three-dimensional point cloud information, and vehicle motion state information comprises vehicle attitude angle, vehicle angular velocity, vehicle linear acceleration and vehicle running speed; The method for establishing the three-dimensional pavement set comprises the steps of setting a vehicle coordinate system as a record Wherein The axle is the forward direction of the vehicle, The axis is the transverse direction of the vehicle, The axle is the vertical direction of vehicle, and laser radar gathers and obtains place ahead road surface point set to: Wherein, the Indicating the number of sample points to be sampled, Represent the first Three-dimensional coordinates of the road points in a vehicle coordinate system; The vehicle state vector is defined as: Wherein, the Is used for the transverse rolling angle, and the transverse rolling angle is used for the transverse rolling angle, In order to be a yaw angle, In order to achieve a roll angular velocity, For the pitch angle rate, In order to achieve a yaw rate, For the longitudinal speed of the vehicle, 、、 The linear accelerations of the vehicle in the longitudinal direction, the transverse direction and the vertical direction, respectively.
3. The unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1 is characterized in that the unified processing in S2 comprises the steps of performing time alignment on various sensor data through a time synchronization algorithm, uniformly converting road point cloud information and vehicle motion state information into a vehicle coordinate system or a global coordinate system through coordinate calibration and coordinate transformation, and performing filtering fusion on multi-source information through a state fusion algorithm to obtain a unified environment state vector; the specific method comprises the steps of setting the laser radar time stamp as IMU time stamp is Wheel speed timestamp is Selecting unified control time Then various types of data are mapped to the same moment by interpolation: Wherein, the And (3) with Respectively sampling time points adjacent to the control time point, The vehicle state estimated value after synchronization; For space unification, the external parameter transformation matrix from the laser radar coordinate system to the vehicle coordinate system is set as follows: Wherein, the In order to rotate the matrix is rotated, For translation vector, laser radar point After being converted into a vehicle coordinate system, the method comprises the following steps: Through the processing, road surface point cloud and vehicle state information under a unified time reference and a unified coordinate system are obtained.
4. The unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1 is characterized in that the three-dimensional geometric modeling in S3 specifically comprises filtering point cloud data and outlier elimination; the specific method comprises selecting a length in front of the vehicle as With a width of Taking the pavement area in the window as a modeling window, and performing quadric surface fitting on the ground points in the window to obtain a local pavement height function model: Wherein, the Indicating the forward direction distance of the vehicle, Indicating the lateral distance of the vehicle, Representing the road surface height of the corresponding point; The curve coefficients to be estimated; The surface coefficients are solved by a least square method, namely, the objective function is minimized: calculating geometric characteristic parameters of the road surface according to the fitting result; The road gradient characteristic parameter calculation method comprises the following steps of Direction and direction The first-order bias of the direction is respectively: the local grade size is defined as: Wherein, the Indicating the point of the road surface Slope magnitude at; the road curvature characteristic parameter calculation method comprises the steps of defining a second-order variation for representing local fluctuation variation speed: Wherein, the Representing the longitudinal curvature component, Representing the transverse curvature component of the lens, Representing a coupling curvature component; the road surface roughness parameter calculating method comprises the steps of setting the actual point cloud height as The height of the fitting curved surface is The local roughness is defined as: Wherein, the The road surface roughness is represented, and the high-frequency fluctuation degree near the fitting plane is reflected.
5. The unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1, wherein the specific method of S4 is that the predicted time domain length is set as The vehicle is at a future time The forward distance reached is approximately: Wherein, the For the future preview location(s), For the current longitudinal speed, Is the current longitudinal acceleration; the positions of the centers of the left wheel and the right wheel of the vehicle in the transverse direction are respectively And Wherein For the wheel track, the preview heights corresponding to the left wheel and the right wheel are respectively as follows: let the equivalent preview position difference in the front-rear wheelbase direction of the vehicle be The front-rear direction height difference is expressed as: the left-right direction height difference is expressed as: thereby establishing pitch and roll disturbance estimators: Wherein, the In order to predict the pitch disturbance angle, To predict roll disturbance angle; Further considering roughness and speed coupling influence, constructing a comprehensive disturbance prediction value: Wherein, the For the roll disturbance to be a predicted amount, Pre-measuring pitch disturbances; 、、、 The disturbance mapping coefficient is obtained through test calibration or system identification; in actual control, a weighted average value in a prediction time domain is selected as an equivalent disturbance input of a current control period: 。
6. the unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1, wherein the specific method of S5 is as follows Calculating disturbance feedforward compensation control quantity, and setting a feedforward gain matrix as follows: the disturbance feedforward control amount is: Wherein, the The feedforward compensation amounts for the roll and pitch directions are shown respectively, 、 Compensating the gain for feedforward; if the disturbance change rate is further considered, the predictive differential compensation term is increased: Wherein, the In order to perturb the rate-of-change gain matrix, The rate of change is predicted for the disturbance.
7. The unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1, wherein the attitude angle control loop in S6: setting the target roll angle and the target pitch angle as respectively 、 The actual roll angle and pitch angle are respectively 、 The attitude error is: the outer loop generates the desired angular velocity command: Wherein, the In order to achieve the desired roll angular velocity, Is the desired pitch rate; 、、 the rolling attitude angle outer ring proportion, integral and differential coefficients are respectively; 、、 the pitch attitude angle outer ring proportion, integral and differential coefficients are respectively; Angular velocity control ring: the actual rolling angular velocity and the pitch angle velocity are respectively 、 The angular velocity error is: the inner ring control output is: Wherein, the 、 Is a feedback control amount; 、、 is the inner ring parameter of the roll angular velocity; 、、 Is the pitch angle rate inner loop parameter.
8. The unmanned vehicle road surface disturbance prediction and active stability control method according to claim 1, wherein the specific method of S7 comprises the following steps of fusing feedforward compensation and feedback control to obtain a final control input: Wherein, the To avoid control saturation, clipping constraints are added: Wherein, the Representing the clipping function, And (3) with The minimum and maximum allowed outputs of the corresponding control channels, respectively.
9. The method for predicting road disturbance and actively stabilizing and controlling an unmanned vehicle according to claim 1, wherein the specific method of S8 is to distribute the attitude control amount to the additional torque of the left and right driving wheels when the vehicle adopts the differential driving mode of the left and right wheels: Wherein, the And The left wheel and the right wheel are respectively added with torque, Distributing coefficients for torque; When the vehicle is equipped with an active suspension, the pitch and roll control amounts are also mapped to suspension damping adjustment amounts: Wherein, the 、 Respectively represents the damping adjustment quantity of the front suspension and the rear suspension, 、 Respectively represents the damping adjustment amounts of the left suspension and the right suspension, 、、、 The coefficients are mapped for the actuator.

Description

Unmanned vehicle road surface disturbance prediction and active stability control method Technical Field The invention relates to the technical field of vehicle stability control, in particular to an unmanned vehicle road surface disturbance prediction and active stability control method. Background With the development of unmanned technology, intelligent perception technology and vehicle motion control technology, unmanned vehicles are widely applied to agricultural operation, logistics transportation, mining area inspection, field detection and other scenes. In such application environments, unmanned vehicles are often required to operate under unstructured or semi-structured road conditions, such as gravel roads, muddy roads, hillside roads, temporary construction roads, and the like. The pavement has obvious irregularity and randomness, and is characterized by large elevation fluctuation, frequent gradient change, uneven local concave-convex distribution, unstable attachment condition and the like. When an unmanned vehicle runs on such a complex road surface, the contact state between the tire and the road surface is continuously changed, and the external excitation generated by the contact state is transmitted to the vehicle body through the chassis structure and the suspension system of the vehicle, so that the fluctuation of pitching, rolling and yawing postures of the vehicle is easily caused. The unstable vehicle posture not only reduces the running smoothness and safety, but also can influence the measurement accuracy of the vehicle-mounted sensor, thereby adversely affecting the environment perception, path planning and task execution. In the prior art, in order to improve the stability of a vehicle, a stable control method based on feedback of a posture state is generally adopted. For example, vehicle attitude angle and angular velocity information is acquired by an inertial measurement unit, and a control input is calculated based on an error between a target attitude and an actual attitude, thereby adjusting a vehicle drive system or a steering system. The control method can achieve a certain effect under the condition of relatively smooth road conditions or relatively small disturbance, but has limitations in complex road surface environments. With the development of vision sensor and laser radar technology, some unmanned vehicles have the ability to sense the road environment in front, such as by recognizing road boundaries through cameras or obtaining three-dimensional point cloud information through laser radar. However, in existing applications, such awareness information is primarily used for obstacle detection or path planning, but is less used in the vehicle stability control process. In other words, the prior art has not fully utilized road surface space geometry to characterize the impending external disturbance to the vehicle. From the vehicle dynamics perspective, road surface irregularities are essentially an external stimulus with spatially distributed characteristics, the effect of which on the vehicle attitude is not only related to road surface height variations, but also to vehicle speed, acceleration, and vehicle structural parameters. If the control adjustment is performed only by relying on the current posture state of the vehicle, the transmission process of disturbance between the environment, the tire, the chassis and the vehicle body is difficult to accurately reflect, so that the control system often lacks foresight. Therefore, in the field of unmanned vehicle stability control, how to effectively combine road environment perception information with a vehicle attitude control process, to establish a road surface disturbance expression mode facing vehicle dynamics response, and to realize disturbance advance compensation in a control system, has become an important technical direction for improving the stable operation capability of an unmanned vehicle under complex terrain conditions. Disclosure of Invention This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application. In order to solve the technical problems, according to one aspect of the present invention, the following technical solutions are provided: a pavement disturbance prediction and active stability control method for an unmanned vehicle comprises the following steps: s1, acquiring road environment information and vehicle motion state, namely acquiring road space information in front of a vehicle and vehicle motion state information in real time, and establishing a three-dimensional road surface set; s2, multi-source information time synchronization and space unification, namely uniformly processing the acquired multi-source data